CAPÍTULO VI - 0

CAPÍTULO VI
ÍNDICE

6.1 Introducción VI-1

6.2 Consideraciones previas VI-2

6.3 Resultados experimentales VI-11

6.4 Criterios de plastificación VI-14

6.4.1 Criterio de Rankine - Lamé VI-14

6.4.2 Criterio de Saint-Venant - Poncelet VI-15

6.4.3 Criterio de Tresca - Guest VI-16

6.4.4 Criterio de Von Mises - Hencky VI-21

6.5 Consideraciones finales VI-31

 CAPÍTULO VI - 1

CAPÍTULO VI
CRITERIOS DE PLASTIFICACIÓN

6-1 Introducción.

De acuerdo con el desarrollo de la programación de este curso, en el primer capítulo se

estudió el estado tensional de los puntos de un sólido. En el segundo capítulo las
deformaciones que sufría pero con la particularidad de que ambos estados se analizaron
de forma independiente. Posteriormente, en el tercer capítulo, se estudió la ley de
comportamiento que une ambos estados. Este proceso de estudio puede ser representado
en el siguiente diagrama:

Objetivo de la
Elasticidad

T u

Lema de Cauchy Relación desplazamiento-

Deformación

σ ij Ley de Hooke ε ij
Ecc. De Lamé

Figura 6-1

Es decir el vector tensión T se relaciona con el tensor de tensiones σij a través del
lema de Cauchy. Luego el tensor de tensiones y el de deformaciones εij se relacionan
entre si a través de la ley de comportamiento, y por último el vector desplazamiento u
se relaciona con el tensor de deformaciones mediante la relación desplazamiento-
deformación. Como puede observarse el objetivo final de la Elasticidad es relacionar el
vector Tensión con los desplazamientos (o lo que es lo mismo carga-desplazamiento) ,
sin embargo esto no puede hacerse directamente sino que es necesario idear unas
variables intermedias, tensiones y deformaciones, y realizar un recorrido a través de
ellas para alcanzar el fin propuesto. Este modelo de la Teoría de la Elasticidad se puede
afirmar que está parcialmente completo a falta, únicamente de añadir algún sistema que
indique cuándo ciertos puntos del sólido abandonan el comportamiento elástico y entran
en fase plástica.

Como se recordará del capítulo III, un sólido entra en fase plástica cuando algun/os de
su/s punto/s superan el Límite Elástico del material. Esta determinación es sencilla de
realizar si el sólido tuviese una geometría muy simple y estuviese sometido a un estado
de cargas igual que al del Ensayo de Tracción. En este caso simplemente con hacer:
 CAPÍTULO VI - 2

σ I ≤σ E

se aseguraría que ningún punto plastificará. En la expresión anterior σE es el Límite

Elástico del material y σI la única tensión existente en el interior del sólido.

Sin embargo lo usual es que un cuerpo tenga una geometría general y esté sometido a
un estado de cargas cualesquiera siendo el caso más general cuando el tensor de
tensiones esté completo. Ante tal situación de cargas cabría preguntarse ¿ Cuándo
alcanza un punto del sólido el Límite Elástico del material ? o bien ¿ Cuál es la
combinación de tensiones más desfavorable que hace que se alcance antes el Límite
Elástico ?.

Para responder a estas preguntas se desarrolla este capítulo y lo que se denomina

Criterios de Plastificación o herramientas que permiten predecir si un punto de un sólido
va a alcanzar la plastificación ante un determinado estado de cargas. A tal fin el capítulo
se articula de la siguiente forma: En primer lugar se hacen unas consideraciones previas,
luego se desarrolla una pequeña revisión de los criterios de plastificación más
relevantes, y por último se hace especial hincapié en los Criterios de Von Mises y de
Tresca aceptados hoy en día por la Instrucción española.

7-2 Consideraciones previas.

Con el fin de sistematizar el establecimiento de un Criterio de Plastificación es

necesario enunciar una serie de postulados. Tales son:

1º Postulado

Si un punto de un sólido elástico plastifica lo hace solo y exclusivamente en función de

las tensiones que actúan sobre él y no de la forma en que las tensiones han alcanzado los
valores críticos. Esta hipótesis permite suponer la existencia de una cierta función
( )
f σ ij (función de plastificación) tal que:

El material se comporta elásticamente si ( )

f σ ij < 0
o bien ( )
f σ ij < 0 6-1
y f (σ ij ) < 0 6-2

Donde ( ) ( )
f σ ij < 0 indica descarga y f σ ij > 0 carga.

De igual forma, un material plastificaría si:

O bien ( )
f σ ij = 0 6-3
Y f (σ ij ) ≥ 0 6-4

La función de función de plastificación f σ ij ( ) se puede escribir genéricamente así:

f = f (σ 11,σ 12 ,σ 13 ,σ 22 ,σ 23 ,σ 33 ) 6-5
 CAPÍTULO VI - 3

Ahora bien si el material es isótropo implica que cualquiera que sea el sistema de
referencia del tensor de tensiones siempre tendrá las mismas tensiones principales. Por
tanto la función f puede ser expresada haciendo uso de las tensiones principales así:

f = f (σ I ,σ II ,σ III ) 6-6

Ahondando aún más en la idea, la función f incluso podría ser expresada, de una forma
más general, en función de los invariantes:

f = f ( I1, I 2 , I 3 ) 6-7

Esta última expresión indica que la función de plastificación ni siquiera depende

directamente de las tensiones principales sino de una combinación de ellas.

2ª Postulado

La función f no debería cambiar si los ejes se intercambian, es decir si el eje 2 pase a

ser el eje 1, por ejemplo, la función de plastificación debería ser la misma. Este hecho
conduce al siguiente resultado:

f = f (σ I ,σ II , σ III ) = f (σ II ,σ I ,σ III ) = f (σ III ,σ II ,σ I ) etc

Una conclusión importante de que la función de plastificación tenga el mismo valor

cualquiera que sea la rotación de ejes es: la función de plastificación f es una función
simétrica de las tensiones principales.

Vamos a extendernos un poco mas y exponer algún ejemplo que ayude a entender este
postulado. A tal fin supóngase el polinomio característico que proporciona las raíces o
tensiones principales:

λ3 − I1 λ2 + I 2 λ − I 3 = 0

Hasta ahora, y por convenio, se ha denominado σ I a la mayor de las tres raíces

cumpliéndose que: σ I > σ II > σ III . Sin embargo una función de plastificación no
puede depender de este convenio, sino que tiene que ser independiente de la forma en
que se denomine a cada raíz o tensión principal. Supóngase dos casos de asignación de
raíces:

Caso 1 Caso 2

σ I = λ1 ; σ II = λ2 ; σ III = λ3 σ I = λ1 ; σ II = λ3 ; σ III = λ2

Está claro que se debe cumplirse que: f = f (σ I ,σ II ,σ III ) = f (σ I ,σ III ,σ II ) .

Si se hace una representación isométrica de las tensiones principales, resulta:

CASO 1 CASO 2
 CAPÍTULO VI - 4

III II

f (σI, σII, σIII ) f (σI, σIII, σII )

σIII σII

σI σIII
σII
σI

I II I III

Figura 6-2

La consecuencia de que f (σ I ,σ II ,σ III ) = f (σ I , σ III ,σ II ) es que f debe ser simétrica

alrededor del eje I, para que al intercambiar σII por σIII la función de plastificación tenga
el mismo valor. Si se repite este razonamiento para los otros dos ejes resulta por
conclusión que la función de plastificación debe ser simétrica respecto a los tres ejes.

3º Postulado

La función de plastificación f toma valores iguales cuando los signos de las tensiones
cambian:

f (σ ij ) = f ( −σ ij ) 6-8

Esto quiere decir que el material se comportará de igual forma tanto si trabaja a tracción
como a compresión, y por tanto la función de plastificación también ha de ser simétrica
respecto a los ejes I , II, III, pero en su parte negativa. Es más, si se requiere que la
condición se mantenga después de que el material haya sido sometido a una
deformación plástica inicial, entonces será necesario no considerar el efecto Baushinger.

Efecto Baushinger:
Sea una probeta de un acero que se somete al siguiente ensayo. Primero se carga y se
sobrepasa el límite elástico (figura 6-2) del material (tramo A B C). Luego se descarga y
se llega a cero. A continuación se invierte la carga y se somete la probeta a compresión
hasta que alcance el límite elástico pero negativo es decir -σE ( tramo C D H). Se
observa que el nuevo límite elástico negativo no coincide con el de tracción ( es menor)
y el comportamiento del material es elástico pero no lineal (efecto Baushinger).
σ
C
B

A D ε

H
- σE

Figura 6-3
 CAPÍTULO VI - 5

4º Postulado

Este último postulado establece que la plastificación sólo depende de la parte desviadora
del tensor de tensiones. Este muy importante postulado es consecuencia de la
observación del hecho experimental (corroborada por los ensayos de Bridgman a
mediados del siglo XX) de que un estado triaxial no provoca plastificación.
Nota:
Cuando una probeta de un material dúctil es sometida a un estado de tracción o
compresión triaxial (un estado triaxial de cargas es tal que todas las presiones sobre un
punto de un sólido son iguales en magnitud y sentido, por ejemplo el estado hidrostático
de la figura 7-3) el sólido no plastifica y la rotura se produce como si el material fuese
frágil (cuando un material se comporta de forma frágil no aparece en ningún instante la
plastificación).

Figura 6-4

Antes de continuar es necesario recordar la descomposición de un tensor en su parte

esférica y desviadora

 I1   I 
 0 0  σ I − 1 0 0 
3   3 
esf desv  I1   I1 
σ ij = σ ij + σ ij = 0 0 + 0 σ II − 0 6-9
 3   3 
 I1   I1 
0 0   0 0 σ III − 
 3  3

I1
Donde = Tσoct , y como puede observarse el tensor esférico crea un estado triaxial de
3
cargas y por tanto no produce plastificación. Consecuentemente el responsable de la
plastificación debería ser la parte desviadora del tensor de tensiones tal y como se
propuso al principio de este postulado.
 CAPÍTULO VI - 6

Representación en el espacio de las tensiones principales

Es una representación que ayuda a entender la superficie de plastificación y consiste en

definir un sistema de referencia cuyos ejes sean las tensiones principales (es importante
llamar la atención que este sistema de referencia no es un sistema cartesiano ni
constituye un espacio por lo que es simplemente una “ representación”). En tal
representación la diagonal principal es una recta en la que se cumple que: σI = σII = σIII
(estado hidroestático) y por tanto es el lugar de los puntos de la parte esférica de un
tensor de tensiones.

σIII

σI = σII = σIII
σijdesv
σij

σijesf
σII

σI

Figura 6-5

También en dicha representación el tensor de tensiones es un punto o un pseudovector

al igual que el tensor esférico y desviador con la particularidad de que son
perpendiculares. La comprobación de tal perpendicularidad consiste en realizar el
producto escalar siguiente:

 I1 I1 I1   I I I  I 
 , ,  ⋅  σ I − 1 , σ II − 1 , σ III − 1  =  1  ⋅ (σ I + σ II + σ III ) −
3 3 3  3 3 3 3
6-10
 I1 I1   I1  I 
 3 ⋅ ⋅  =   (I1 ) −  1 ⋅ I1  = 0
 3 3 3 3 

y ver que efectivamente es nulo

Realizadas las consideraciones anteriores, el siguiente paso es definir la superficie de

plastificación como un cilindro cuyo eje es concéntrico a la diagonal principal del
pseudosistema mencionado. A esta conclusión se llega observando que la plastificación
no depende de la posición de σ ijesf sino de σ ijdesv y por tanto cualquier sección
realizada por un plano perpendicular a la diagonal principal debe tener la misma forma
tal y como muestra la figura siguiente:
 CAPÍTULO VI - 7

f ( σI , σII ,σIII )

σIII

σijdesv

σijesf
σII

σI

Figura 6-6

Representación de Haig – Westergaard

Consiste en realizar una intersección a la superficie f por un plano perpendicular a la

diagonal principal y que pase por el origen. El resultado es una representación plana de
los infinitos pseudovectores desviadores (hay que recalcar que la intersección es
independiente del punto sobre el que se realice ya que la superficie f es un cilindro). Si
se elige un sistema isométrico para representar el resultado de la intersección se obtiene
la figura siguiente:

III
Superficie f de
Plastificación
intersectada por un
plano

I II

figura 6-7

La aplicación del segundo postulado a la función de plastificación conduce a que debe

ser simétrica alrededor de los ejes (I ,II ,III), y si no se considera el efecto Baushinger
también debería ser simétrica alrededor de los ejes ( I, II, III) pero en su parte negativa.
Esta condición obliga a que la función de plastificación sea simétrica respecto a 6
segmentos de área de 60º cada uno, tal y como se muestra en la figura 6-8 (las distancias
OM y OQ son constantes).
 CAPÍTULO VI - 8

III

Q Simétrica
Simétrica

3 2 M Simétrica
M

4 1
O

5 Q
Simétrica Q 6
Simétrica
I M
II
Simétrica

Figura 6-8

La simetría obliga a que el segmento de área 2 deba ser imagen especular del segmento
de área 1; igualmente el 3 debe ser la imagen especular del 2, y así sucesivamente.
Consecuentemente la función de plastificación ha de ser simétrica respecto a unos ejes a
30º ( figura 6-9). Estos condicionantes de simetría restringen la forma de la función de
plastificación en el plano de Haig-Westregaard ya que además de ser simétrica respecto
a los ejes I; II; y III también debe ser simétrica respecto a cada segmento de área
contiguo.
III

II
I

I II

III

Figura 6-9

La forma dibujada de estrella es simplemente para mostrar un caso general de una

posible forma de la función de plastificación que cumple los requisitos de simetría,
aunque no parece lógico que un sólido tenga tal comportamiento en la plastificación.
Sin embargo la forma que se representa en la figura 6-10 también es válida, en cuanto a
requisitos de simetría, pero presenta un comportamiento más lógico de plastificación
tendiendo hacia la forma de circunferencia que, por supuesto, también cumpliría con las
condiciones impuestas.
 CAPÍTULO VI - 9

III

II
I

I II

III

Figura 6-10

Otro comportamiento también válido, y muy parecido al real, sería como muestra la
figura 6-11

III

II
I

I II

III

Figura 6-11

Por último hay que añadir que existen infinitas posibilidades para descomponer un
tensor en uno esférico y otro desviador. Sin embargo sólo existe una que hace que los
pseudovectores sean perpendiculares y es tomar el esférico de tal forma que sus tres
tensiones principales sean precisamente las tensiones octaédricas.

6-3 Resultados experimentales

Cualquier criterio de plastificación que se obtenga de una forma más o menos teórica
deberá estar de acuerdo, o bien sus predicciones deberían acercarse o estar próximas, a
los resultados obtenidos de forma experimental. Lo ideal sería que en una situación
compleja de carga y geometría se pudiese realizar un ensayo que proporcionase el
verdadero comportamiento y Límite Elástico del caso que está estudiando.

La realidad, sin embargo, es que los ensayos se tienen que realizar con sólidos de
geometría sencilla y estados de carga simples y luego extrapolar las consecuencias a
situaciones más complejas a través de los criterios de Plastificación.
 CAPÍTULO VI - 10

Los ensayos que normalmente se emplean, aparte del ensayo de tracción, tratan de
simular situaciones complejas, para ello se usan tubos y se les somete a cargas axiles
de tracción, torsión, y presión. Combinando estos tres tipos de carga se consiguen una
serie de ensayos donde aparecen tensiones normales y tangenciales, dos tensiones
normales, etc. Por ejemplo en la figura 6-12 se muestra un tubo de pared delgada y los
resultados teóricos que se obtienen cuando es sometido a axil, torsor, y presión interna:

σzϕ T
σ1
T

σϕϕ
e P

R
F

Figura 6-12

Las tensiones que provocan este tipo de esfuerzos son:

F
- Debido al axil F : σ 1 =
2π r e
T
- Debido al Momento Torsor T: σ z ϕ =
2π r 2 e
R
- Debido a una presión interna p suponiendo tapaderas móviles: σ ϕϕ = p
e

Por tanto se dispone de dos tensiones normales: σ 11 y σ ϕϕ y de una tensión

tangencial: σ 1ϕ que se pueden combinar produciendo estados de carga cuasi-planos.
Los casos extremos son: que sólo exista axil, o que sólo exista torsión:

Tτ Tτ

F ≠ 0 ; T=0 ; p = 0 F = 0 ; T ≠0 ; p = 0

σII = 0 σI = σ11 σI = - σz ϕ σII = 0

σIII =0 σI = σz ϕ Tσ
Tσ

Figura 6-13
 CAPÍTULO VI - 11

y el objetivo de los ensayos es obtener resultados experimentales de las combinaciones

de ambos estados de carga.

Entre los mas conocidos están los realizados por Lode publicados en 1926, Ros y
Eichinger publicados en 1926 y los de Taylor y Quinney publicados en 1931. Los
ensayos tenían por objetivo corroborar los criterios de plastificación teóricos con los
obtenidos experimentalmente utilizando para ello tres tipos de materiales: acero, níquel,
y cobre. Los resultados mostraron que existía una buena correspondencia entre los
valores teóricos y los obtenidos experimentalmente si se empleaba el criterio de Von
Mises- Hencky y bastantes próximos si se empleaba el criterio de Tresca, Los resultados
de los autores Taylor y Quinney usando tubos de cobre, aluminio y acero y
sometiéndolos a carga axil y momento torsor se representan en la figura 6-14 que
muestra la variación del Límite Elástico en función de las tensiones normales y
tangenciales.

Figura 6-14

Como puede observarse los resultados se expresan en unos ejes adimensionales. Ello
significa que, teóricamente, cualquier material debería plastificar siempre en el mismo
punto para una determinada combinación de axil y cortante, o lo que es lo mismo que
los símbolos círculo, triángulo y cuadrado de la figura 6-14 deberían coincidir para
cualquier punto. Lógicamente la realidad no es así, pero puede observarse que es
bastante parecida.
T 
Los extremos de la batería de ensayos son los casos de axil puro  σ = 1 y el ensayo
σE 
T 
de torsión pura  τ = 1 . Según muestran los resultados en un caso de torsión pura
σE 
(sólo existiría tensión tangencial) se obtiene:

Tτ
= 0.56 6-11
σE

Este resultado es importante porque va a servir de test o prueba para comprobar la

exactitud o inexactitud de un criterio de plastificación cuando se particularice para el
caso de torsión pura.
 CAPÍTULO VI - 12

6-4 Criterios de Plastificación

El objetivo fundamental de un criterio de plastificación es proveer al diseñador de una

ley o regla que le informe, de una forma sencilla, si las tensiones en un punto de un
sólido van a alcanzar un límite establecido. En nuestro caso es el Límite Elástico del
material. Históricamente se trataba de ligar directamente los resultados del ensayo de
tracción con los que se obtenía del cálculo. Así surgieron criterios como los Rankine y
Saint-Venant, que siendo lógicos para su época hoy en día son de poco uso. Sin
embargo parece instructivo exponerlos, y así se tendrá una idea de como ha sido la
evolución del conocimiento en este campo.

6-4-1 Criterio de Rankine-Lamé (1858 y 1852)

Este criterio es el más inmediato, y establece que la plastificación se producirá cuando

las tensiones existentes en algún punto del sólido alcancen el Límite Elástico σE del
material. Es decir:

σ I ≤ σ E ; σ II ≤ σ E ; σ III ≤ σ E 6-12

Este criterio se representa en el espacio de las tensiones principales como un cubo de

lado 2 σE centrado en el origen, figura 6-15-a, y en el espacio de Haig-Westergaard,
figura 6-15-b, sería la intersección del cubo por un plano perpendicular a la diagonal
principal pasando por el origen.

σIII Intersección del cubo

con un plano que pasa
por el origen

(-σE , σE , σE )

σII
(-σE , σE , -σE )

σI (σE , σE , -σE )
√2σE

Figura 6-15

Este criterio no es válido para materiales dúctiles ya que para el ensayo siguiente:
 CAPÍTULO VI - 13

Tτ
σ
•

σ σ - σII≡σ σI≡σ Tσ

•
σ

Figura 6-16

predeciría que la plastificación se producirá cuando la tensión tangencial Tτ = σ = σ E (

T
en realidad debería expresarse que τ = 1 ). Este resultado es casi el doble del
σE
obtenido mediante los ensayos de Lode, o Quinney, que dice que la plastificación se
T
producirá cuando τ = 0.56 . Consecuencia de ello es que este criterio no es válido
σE
para materiales dúctiles. Sin embargo sí es un criterio válido para materiales frágiles,
por ejemplo en rotura de cristales, ya que en este tipo de materiales si se cumple que
Tτ = σ I = σ E .

6-4-2 Criterio de Saint-Venant -Poncelet (1870 y 1839)

Este criterio, propuesto independientemente por ambos autores, establece que un punto
de un sólido plastificará cuando las deformaciones principales alcancen el valor de la
deformación ε E (deformación correspondiente al Límite Elastico en el ensayo de
tracción).

Puesto que:
σE
εE =
E

la plastificación ocurrirá (en un caso general) cuando:

ε I ≤ ε E ; ε II ≤ ε E ; ε III ≤ ε E 6-13

Si se expresa las deformaciones en función de las tensiones a través de la ley de Hooke:

εI =
1
[σ I − v (σ II + σ III ) ] = σ E 6-14
E E

o bien:
 CAPÍTULO VI - 14

[σ I − v (σ II + σ III ) ] = σ E 6-15

y de igual forma para las otras direcciones:

[σ II − v (σ I + σ III ) ] = σ E 6-16
[σ III − v (σ II + σ I ) ] = σ E 6-17

La comprobación de la bondad de los resultados se realiza para el caso de carga del

criterio anterior. Efectivamente σ I = σ σ II = −σ y σ III = 0 . Sustituyendo estos
valores en 6-15 ; 6-16 ; y 6-17 resulta:

[σ I − v (σ II + σ III ) ] → (1 + v )σ ≤ σ E
[σ II − v (σ I + σ III ) ] → −(1 + v)σ ≤ σ E
[σ III − v (σ II + σ I ) ] → 0 ≤ σ E
σE
El más desfavorable es: σ ≤ y debido a que (Tτ )max = σ resulta:
1+ v
σE
Tτ ≤ 6-18
(1 + v)

Tomando un valor del módulo de Poisson de 0.30 resultaría:

Tτ 1 1
= = = 0.77 6-19
σE 1 + v 1.3

Valor que, mejorando el obtenido con el criterio anterior sigue aún estando alejado del
T
establecido experimentalmente ( τ = 0.56 para un caso de cortadura pura).
σe

En el espacio de la tensiones principales este criterio representaría un romboedro, y la

intersección con el plano perpendicular a la diagonal principal sería un hexágono
regular Figuras 6-17-a y 6-17-b.

Figura 6-17
 CAPÍTULO VI - 15

6-4-3 Criterio de Tresca - Guest (1868 y 1872)

Durante la realización del ensayo de tracción se observa que en las inmediaciones del
Límite Elástico aparecen en el material unas líneas a 45º, conocidas como líneas de
Lüders, que hicieron pensar la posibilidad de que la plastificación estuviese gobernada
por las tensiones tangenciales (como se sabe las tensiones tangenciales máximas se
producen en planos que forman 45º con la dirección principal I ).

El criterio de Tresca establece que: un punto de un sólido plastificará cuando la tensión

tangencial máxima Tτ alcance el valor de la tensión tangencial TτE . Siendo TτE la
tensión tangencial que se produce en el ensayo de tracción cuando la probeta alcanza el
límite elástico del límite elástico en el ensayo de tracción.

El circulo de Mohr del ensayo de tracción es:

TσE

σE

Figura 6-18
σE
Y por tanto TτE =
2
Para el caso general de un punto de un sólido en tres dimensiones, los círculos de Mohr
correspondientes serían:

Tτ

( Tτ )max

σIII σII σI Tσ

Figura 6-19
 CAPÍTULO VI - 16

σ I − σ III
y la tensión tangencial máxima es: Tτmax =
2

Para un caso general, como el anterior, y no teniendo en cuenta el convenio de que

σI > σII > σIII el criterio de Tresca se expresaría de la siguiente forma:

σ I − σ III σE σ I − σ II σE σ II − σ III σE
= ; = ; = 6-20
2 2 2 2 2 2

o bien:

σ I − σ III = σ E ; σ I − σ II = σ E ; σ II − σ III = σ E 6-21

A continuación es necesario hacerse la siguiente pregunta ¿ Cuánto de bien, o cuánto de

mal, funciona este criterio?

Para ello se retoma el ensayo de cortadura pura: σ I = σ σ II = −σ ; σ III = 0

Sustituyendo estos valores en 6-21 resulta:

σ I − σ III ≤ σ E ⇒ σ ≤ σ E 6-22
σ I − σ II ≤ σ E ⇒ 2σ ≤ σ E 6-23
σ II − σ III ≤ σ E ⇒ − σ ≤ σ E 6-24

σE
De las desigualdades anteriores la mas desfavorable es σ ≤ . Por tanto el resultado
2
que predice el criterio de Tresca para este tipo de caso es que la plastificación se
producirá cuando la tensión tangencial alcance el valor de:

(σ = (Tτ )max ) ≤ σ E 6-25

2

y adimensionalizando:

Tτ
= 0.5 6-26
σE

este valor teórico (0.5) es muy próximo al experimental (0.56) concluyéndose que el
criterio de Tresca produce resultados bastante aceptables.

En el espacio de las tensiones principales cada una de las igualdades de la expresión 6-

21 representa a dos planos paralelos a la diagonal principal de acuerdo con la figura 6-
20:
 CAPÍTULO VI - 17

σ III

σ II

σI
Figura 6-20

No es inmediato vislumbrar geométricamente tal conclusión. Para ello se considera uno

de los casos por ejemplo: σ I − σ II = σ E y se supondrá que el resultado es positivo.

Está claro que en el pseudoespacio de las tensiones principales σ I − σ II = σ E es un

plano tal que corta al eje σI en σE y al eje σII en - σE (puntos A y B de la figura 6-21) y
es paralelo al eje σIII (misma figura).

Por otra parte se sabe que la diagonal principal forma ángulos iguales con los tres ejes
principales de valor: arccos( 1 / √ 3 ) que a su vez se puede expresar como:

 1   2
arccos   = arcsen  

 3   3 
σIII

Diagonal Principal
Plano σI - σII = σE

O
B• σII
 2
• arcsen  

A 3
 

σI

figura 6-21
Si se hace el producto escalar del vector normal al plano σ I − σ II = σ E y el vector
coincidente con la diagonal principal y el resultado fuese nulo significaría que ambos
son perpendiculares, o lo que es lo mismo que el plano σ I − σ II = σ E es paralelo a la
diagonal principal. Para ello:
 CAPÍTULO VI - 18

 2 2 
Vector normal al plano σ I − σ II = σ E =>  ,− ,0 
 2 2 
 1 1 1 
Vector en la dirección de la diagonal principal:  , , 
 3 3 3
realizando el producto escalar:

 2 2   1 1 1 
 
 2 ,− 2 ,0  ⋅  3 , 3 , 3  = 0
 

Lo cual corroborara la aseveración de que el plano σ I − σ II = σ E es paralelo a la

diagonal principal. De igual forma se podría hacer para el resto de los planos y
comprobar que efectivamente son todos paralelos a la diagonal principal y ratificar la
veracidad de la figura 6-20.

Para hacer una representación de Haig-Westergaard se intersecta el cilindro de Tresca

con un plano perpendicular a la diagonal principal resultando un hexágono regular, que
( )
se representa en la figura 6-22. Los cuyos ejes se han denominado σ ´I , σ ´II , σ ´III para
no confundirlos con los principales y se han situado de acuerdo con una representación
isométrica.

σIII´

σII - σIII = -σE σIII - σI = σE

σI - σII = σE σI - σII = -σE

O

A´

σI´ σIII - σI = -σE σII´

σII - σIII = σE

2 2
OA´ = OA = σE
3 3

Figura 6-22

Es interesante remarcar que en la figura 6-22 la distancia O A´ no coincide con la

distancia OA del espacio de las tensiones principales puesto que son representaciones
2
diferentes y para demostrar que OA´ = OA se hace uso de la figura 6-23. En ella
3
se representa un plano perpendicular a la diagonal principal sobre el que se proyecta el
segmento OA resultando:
 CAPÍTULO VI - 19

σIII
Plano
perpendicular a Diagonal
σ´III la diagonal Principal
principal

σ´II
O
σII
α
γ
A´
• • β = 90º
A

σI´ α = arcsen(√2 / √3)

σI

Figura 6-23

El segmento OA´es la proyección del segmento OA sobre el eje σI´ y se cumple que:

γ = 90 − α

2
Por tanto: OA´= OA cos γ = OA sen α = OA tal y como se quería demostrar.
3

Un tema interesante es que el criterio de Tresca puede ser escrito en función de los
invariantes de la siguiente forma:

2 4 6
σ σ σ
4 I 23 − 27 I 32 − 36 I 22 e + 96 I 2 E − 64 E = 0
4 16 64

La importancia de esta expresión radica en poner de manifiesto dos cosas:

a) que el criterio de Tresca es el primero en exponer que es independiente del primer

invariante y por consiguiente independiente del tensor esférico.

b) Por otra parte, al ser expuesto en forma de invariantes significa que es independiente
de la orientación del sistema de referencia y de la numeración de este.

En los casos bidimensionales en que σ II es nula el criterio se expresa particularizando

la expresión 6-10 para: σ I = σ I ; σ III = σ III ; σ II = 0 . Resultando:

σ I − σ III σE σI σE σ III σE
= ; = ; = 6-27
2 2 2 2 2 2

La representación de estas condiciones en el plano ( σI , σIII ) es:

 CAPÍTULO VI - 20

σIII

σIII = σE

σIII - σI = σE σI = σE

σI

σI - σIII = σE
σI = - σE

σIII = - σE

Figura 6-24

7-4-3 Criterio de Von Mises - Hencky (1913 y 1924)

Tratando de suavizar la superficie angulosa de plastificación que prevé el criterio de

Tresca, Von Mises propuso, de forma heurística, que la representación en el espacio de
las tensiones principales de un criterio de plastificación debería ser un cilindro de
2
sección circular de radio: σ E en vez de un cilindro de sección hexagonal como en
3
el caso de Tresca. Tal representación sería:

Figura 6-25

Y la representación de Haig-Westergaard:
 CAPÍTULO VI - 21

σ´III

Von Mises

Tresca
O

A´

σ´I σ´II

2
OA´ = σE
3

Figura 6-26

La demostración teórica del criterio fue realizada posteriormente por Hencky. Por
último Naday en la década de los años 30 puso de manifiesto la coincidencia con las
tensiones tangenciales octaédricas.

El criterio de Von Mises - Hencky se puede expresar de la siguiente forma: Un punto de

un sólido plastificará cuando la densidad de energía de distorsión (densidad de energía
asociada al tensor desviador) alcance el valor de la densidad de energía de distorsión
que tiene una probeta en un ensayo de tracción en el instante en que la tensión llegue al
límite elástico.

La Densidad de Energía de Deformación (energía por unidad de volumen) almacenada

durante el proceso de carga de un sólido cualquiera es:

1
U = σ ij ε ij
2

Sustituyendo las deformaciones en función de las tensiones, se obtiene:

U = 
( 2 3
)
1  σ 11 + σ 22 + σ 33 − 2v (σ 11σ 22 + σ 222σ 33 + σ 11σ 33 ) + 
2

( )
6-28
2 E 2 (1 + v ) σ 2 + σ 2 + σ 3 
 12 23 13

y en el caso de que el tensor esté expresado en el sistema principal sería:

U =
1
2E
[(
σ I2 + σ II2 + σ III
3
)
− 2v (σ I σ II + σ II σ III + σ I σ III ) ] 6-29

Por otra parte un tensor de tensiones expresado en el sistema principal y descompuesto

en su parte esférica y desviadora es:
 CAPÍTULO VI - 22

 I1   I 
 0 0  σ I − 1 0 0 
3   3 
esf desv  I1   I1 
σ ij = σ ij + σ ij = 0 0 + 0 σ II − 0
 3   3 
 I1   I 
0 0   0 0 σ III − 1 
 3  3

La Densidad de Energía de Deformación sería la suma de la Densidad de Energía de

Deformación de la parte esférica, más la correspondiente a la parte desviadora, así:

U = U esf + U desv 6-30

Puesto que en la plastificación sólo interviene la parte desviadora se halla a

continuación la Densidad de Energía de Deformación correspondiente al tensor
desviador. Su obtención es simplemente sustituyendo en 6-29:

 I1   I1   I1 
σ I → σ I −  ; σ II →  σ II −  ; σ III →  σ III − 
 3  3  3

resultando:

1  I1  
2 2 2
desv I1   I1  
U =   σ I −  +  σ II −  +  σ III −   −
2E   3  3  3 
  6-31
v  I  I   I  I   I  I 
  σ I − 1   σ II − 1  +  σ II − 1   σ III − 1  +  σ I − 1   σ III − 1 
E  3 3  3 3  3 3 

Operando:

U desv =
1

2
σ 2 + σ 2 + σ 2 − σ + σ + σ ( 2
) ( σ I + σ I + σ )2
I −
I II III I +
2E  3 I I
3 
 
6-32
 I I I I  
2
v  − 1 (σ I + σ II ) − 1 (σ II + σ III ) − 1 (σ I + σ III ) + 3  1  + 
3 3 3 3 
E
σ σ + σ σ + σ σ 
 I II I III II III 

Sumando y restando a cada paréntesis del segundo corchete lo que le falta para
completar el primer invariante, se obtiene:
 CAPÍTULO VI - 23

U desv
=

1  2 2 2
σ I + σ II + σ III −
σI +σI +σI  ( 2
−
)
2E  3 
 
6-33

v  σI +σI +σI
−
( 2
) 
+ σ I σ II + σ I σ III + σ II σ III 
E  3 
 

Multiplicando y dividiendo por 3 y operando se obtiene:

U desv =
1
6E
[
2σ I2 + 2σ II
2 2
+ 2σ III − 2σ I σ II − 2σ I σ III − 2σ II σ III + ]
6-34
v
6E
[
2σ I2 + 2σ II
2 2
+ 2σ III − 2σ I σ II − 2σ I σ III − 2σ II σ III ]
expresión que representada en su forma habitual es:

U desv =
1+ v
6E
[ (σ I − σ II )2 + ( σ I − σ III )2 + ( σ II − σ III )2 ] 6-35
o bien:

U desv =
1
12 G
[ (σ I − σ II )2 + ( σ I − σ III )2 + ( σ II − σ III )2 ] 6-36

La Densidad de Energía de Deformación de la parte desviadora del tensor de tensiones

en el ensayo de tracción cuando se alcanza el límite elástico se obtiene particularizando
la expresión anterior para: σI = σE ; σII = 0 ; σIII = 0 obteniéndose:

desv
U ens tracc =
1
12 G
[ (σ E)
2
+ ( σ E )2 ] 6-37

Igualando se obtiene el conocido criterio de Von Mises:

[ (σ I − σ II )2 + ( σ I − σ III )2 + ( σ II − σ III )2 = 2σ E2 ] 6-38

o bien:

( σ I − σ II )2 + ( σ I − σ III )2 + ( σ II − σ III )2
≤σ E 6-39
2

Como puede observarse, el criterio de Von Mises representa claramente un cilindro de

sección circular cuyo eje es la diagonal principal tal y como se representa en la figura 6-
25. La intersección con el plano σ III = 0 es una elipse que si se proyecta en el plano
2
perpendicular a la diagonal principal es un cilindro de radio σ E de acuerdo con la
3
figuras 6-25 y 6-26.
 CAPÍTULO VI - 24

Existen varias formas de obtener el mismo resultado, por ejemplo:

- Limitando el segundo invariante del tensor desviador. Por tanto el criterio

también podría definirse como: La plastificación de un punto de un sólido comenzará
cuando el segundo invariante del tensor desviador alcance el valor del correspondiente
al del ensayo de tracción.

- Limitar la tensión intrínseca tangencial octaédrica del tensor desviador. Por tanto
también podría definirse como: La plastificación de un punto de un sólido elástico
comenzará cuando la tensión intrínseca tangencial octaédrica del tensor desviador
alcance la tensión intrínseca tangencial octaédrica del tensor desviador correspondiente
al ensayo de tracción .

En el caso bidimensional ( σII = 0 ) el criterio de Von Mises adopta la siguiente

expresión:

( σ I )2 + ( σ I − σ III )2 + ( σ III )2 ≤ σ 6-40

E
2

o bien:

σ I2 + σ III
2
− σ I σ III ≤ σ E2 6-41

que es la ecuación de una elipse. La figura siguiente representa la elipse del criterio de
Von Mises superpuesta al hexágono irregular del criterio de Tresca:

σIII

σI

Figura 6-27

Por último la comprobación de la fiabilidad de este criterio frente a un caso real se

realiza a través del ensayo de cortadura. Es decir

σI = σ ; σ II = −σ ; σ III = 0 6-42

sustituyendo se obtiene:
 CAPÍTULO VI - 25

(2 σ )2 + 2 ( σ )2 ≤ σ E → 3σ 2 ≤ σ E → σ ≤
σE
6-43
2 3

(Tτ )max
Como en este ensayo (Tτ )max = σ resulta: (Tτ )max ≤ σ E → ≤ 0.577
3 σE

Valor muy parecido al obtenido experimentalmente ( 0.56) , concluyéndose que este

criterio es el que proporciona unos resultados teóricos más próximos a los
experimentales.

Para terminar es interesante expresar el criterio de Von Mises desde el punto de vista de
la Resistencia de Materiales. En esta disciplina la forma usual del tensor de tensiones
es:

 σ 11 σ 12 σ 13 
 
σ ij =  σ 12 0 0  6-44
σ 0 0 
 13

ya que supone que las tensiones σ22 , σ33 y σ23 tienen valores despreciables. Si se
hallan las tensiones principales se obtiene:

σI =
2
σ 11 + σ 11 (2
+ 4 σ 12 2
+ σ 13 ) σ II = 0 σ III =
2
σ 11 − σ 11 (
2
+ 4 σ 12 2
+ σ 13 )
2 2

Y sustituyendo estos valores en el criterio de Von Mises se obtendría:

2
σ 11 ( 2
+ 3 σ 12 2
)
+ σ 13 ≤ σE 6-45

y en el caso muy común de que sólo existe una tensión normal ( σ ) y una tensión
tangencial ( τ ) resulta:

σ 2 + 3 (τ )2 ≤ σ E 6-46

A continuación se muestra un gráfico de los resultados de los ensayos experimentales

obtenidos por Taylor y Quinney y se comparan con los que proporcionaría el criterio de
Tresca y de Von Mises. El ensayo consistió en someter a un tubo (de pared delgada) a
una combinación de dos acciones: un esfuerzo axil de tracción y un momento torsor. El
esfuerzo axil produce un tensor de tensiones donde sólo existe σ I mientras que el
momento torsor produce un tensor de tensiones donde sólo existe σ12 . Por tanto se está
en un caso que corresponde a la última expresión. Combinando las acciones se produce
una nube de puntos y la conclusión es evidente: el criterio de Von Mises proporciona
mejor aproximación a los resultados del ensayo que cualquier otro.
 CAPÍTULO VI - 26

Figura 6-28

7-5 Consideraciones finales.

Hasta ahora se ha hablado de superficies de plastificación, pero no de superficies de

roturas. El tema tiene su lógica ya que un material deja de comportarse elásticamente
cuando alcanza el límite elástico y dejan de ser válidas las ecuaciones de la elasticidad
cuando lo supera de ahí la importancia de su determinación. Por otra parte la
determinación de la superficies de plastificación solo tiene sentido en materiales
dúctiles pero no en materiales frágiles.

Inmediatamente surge la cuestión: ¿ Cuándo un material es dúctil o frágil? . Esta

pregunta trató de ser respondida en el capítulo concerniente al estudio de la ley de
comportamiento. Sin embargo hay que tener presente que no siempre un material dúctil
se comporta como tal ya que si trabajara en forma triaxial tendría un comportamiento
frágil, no siendo de aplicación lo visto en este capítulo. Por ello es muy importante que
situaciones de carga que conducen a estados triaxiales sean, en la medida de lo posible,
soslayados mediante disposiciones constructivas adecuadas. Es el caso que se presenta
a continuación:
3

Cordones de
soldadura

Figura 6-29

En esta figura se representan tres piezas de acero unidas por tres cordones de soldadura
formando ángulos de 90º entre sí. La realización de un cordón de soldadura genera en el
material unas tensiones (de origen térmico y producidas por el calor generado en el
proceso de soldeo), que permanecen en él y se denominan tensiones residuales. Así:

El cordón de soldadura sobre el eje 1 producirá σ11

 CAPÍTULO VI - 27

El cordón de soldadura sobre el eje 2 producirá σ22

El cordón de soldadura sobre el eje 3 producirá σ33

Consecuentemente el origen estará sometido a un estado triaxial de cargas no deseable,

por ello es necesario buscar alternativas constructivas que eviten estas situaciones,
siendo una de ellas ejecutar los cordones de la siguiente forma:

Cordones de
soldadura
Hueco

Figura 6-30

De esta forma el origen estará libre de tensiones y no se producirá un estado triaxial de

cargas.
 CAPÍTULO VI - 28

William John Macquorn Rankine

Born: 2 July 1820 in Edinburgh, Scotland

Died: 24 Dec 1872 in Glasgow, Scotland

Trained as a civil engineer, William Rankine was appointed to the chair of civil
engineering and mechanics at Glasgow in 1855. He developed methods to solve the
force distribution in frame structures.
He worked on heat, and attempted to derive Sadi Carnot's law from his own hypothesis.
His work was extended by Maxwell. Rankine also wrote on fatigue in the metal of
railway axles, on Earth pressures in soil mechanics and the stability of walls. He was
elected a Fellow of the Royal Society in 1853.
Among his most important works are Manual of Applied Mechanics (1858), Manual of
the Steam Engine and Other Prime Movers (1859) and On the Thermodynamic Theory
of Waves of Finite Longitudinal Disturbance.
 CAPÍTULO VI - 29

George Biddell Airy

Born: 27 July 1801 in Alnwick, Northumberland, England
Died: 2 Jan 1892 in Greenwich, England

George Airy's father was William Airy while his mother was Ann Biddell. William Airy
was from Lincolnshire and Ann was the daughter of a farmer from Suffolk. Originally
William had been a farmer too, but he had educated himself and risen to the position of
tax inspector. When George was born his parents were living in Northumberland where
William was a collector of excise, but in the following year the family moved to
Hereford when William was transferred there.
George attended Byatt Walker's school in Colchester and at the age of ten he took first
place at the end of his primary school career. He had learnt some useful skills at the
school such as arithmetic, double-entry book keeping and how to use a slide rule. He
had probably learned more, however, from studying his father's books. He wrote in his
autobiography that (see [3]):-
... he was not a favourite with his school mates.
Eggen writes [1]:-
An introverted but not shy child, Airy was, even for the time and especially for his
circumstances, a young snob. Nevertheless, he overcame some of the dislike of his
schoolmates by his great skill and inventiveness in the construction of peashooters and
other such devices.
Before Airy left Byatt Walker's school his father had transferred again, this time to
Essex. From 1812 Airy spent his summers with his uncle, Arthur Biddell, who had a
farm near Ipswich. Clearly Airy was not too happy at home because he asked his uncle
if he could live with him rather than with his own family. Things had taken a turn for
the worse at home since his father lost his tax collectors job in 1813 and the family
were, from that time, living in poverty. Because of the financial circumstances the
family seem to have been quite glad that Airy's uncle had almost taken over the role of
his father.
The fact that Airy spent about half his time with his uncle over the next five years was
important for him. Arthur Biddell was a man of learning who had a fine library
containing books on chemistry, optics and mechanics which Airy avidly studied, and in
addition he had many leading scientists as his friends. Their influence on the young
Airy was marked and was a major factor in his seeking an academic career.
During these five years, 1814 to 1819, Airy attended Colchester Grammar School where
he was [7]:-
... noted for his memory, repeating in one examination 2394 lines of Latin verse.
Airy entered Trinity College, Cambridge in 1819 as a sizar, meaning that he paid a
reduced fee but essentially worked as a servant to make good the fee reduction.
 CAPÍTULO VI - 30

However it was only because his uncle provided financial support that he was able to
undertake university studies at all. To supplement his income Airy took private pupils
and this, of course, gave him less time for his own studies. Despite this his performance
was outstanding and he graduated as Senior Wrangler (the top First Class student) in
1823 and was a Smith's prizeman. Woodhouse, who had left the Lucasian chair in 1822
to become Plumian Professor of Astronomy, was one of Airy's examiners for the
Smith's prize, the other being Thomas Turton who had succeeded Woodhouse to the
Lucasian chair. In the following year Airy was awarded a fellowship at Trinity College
and began his academic career.
We should comment on why Airy did so well in the Tripos examinations, being far
ahead of the next best student. The Tripos examinations at that time were less a test of
mathematical ability and more a test of the candidates ability to learn vast amounts of
material and methods. At this Airy proved exceptionally good, partly because of his
excellent memory, but also because of his remarkable organisational abilities. As an
undergraduate he kept paper beside him to record every thought he had. Later
everything was transferred to the books and diaries which he kept. He maintained this
routine throughout his life and this record, almost of his every thought, still exists to
provide remarkable evidence of the period [3]:-
The ruling feature of his character was order. From the time he went up to Cambridge to
the end of his life his system of order was strictly maintained.
Clerke writes in [7]:-
He never destroyed a document, but devised an ingenious plan of easy reference to the
huge bulk of his papers.
In 1824 Airy met Richarda Smith while on a walking holiday. He proposed two days
after they first met but her father, Richard Smith, the vicar of a church near Chatsworth,
refused to allow the marriage on the grounds that Airy could not support his daughter
financially. This made Airy determined to obtain a position with the financial status
which would allow him to marry.
Only three years after graduating from Cambridge, he was appointed Lucasian Professor
of Mathematics at Cambridge. It is rather surprising that the Lucasian Professor only
received 99 per year while Airy was already receiving 150 as an assistant tutor. Airy
wondered whether he could afford to compete for the chair when he was advised in
1826 that Turton was leaving, but Peacock persuaded him that the status was more
important than the money. He became one of three candidates, French and Babbage
being the other two. When Babbage stated that he was about to start legal proceedings
over the election, French withdrew. Airy triumphed and a rivalry with Babbage which
was to last for many years began.
In addition to the Lucasian Chair, Airy was appointed a member of the Board of
Longitude which gave him another 100 per year provided he attended four meetings.
He explained his actions (see [3]):-
My prospects in the law or other profession might have been good if I could have
waited but marriage would have been out of the question and I much preferred a
moderate income in no long time. I had now in some measure taken science as my line
(but not irrevocably) and I thought it best to work it well for a time at least and wait for
accidents.
These, of course, are not the words of a man driven by a love of his subject. He
certainly still did not have the financial position to allow him to marry Richarda so he
tried for other posts. His attempt to secure the vacant post of Astronomer Royal for
Ireland failed in 1827.
 CAPÍTULO VI - 31

Airy was an examiner for the Smith's Prize and gave lectures while holding the
Lucasian Chair. He lectured on light and in these lectures he explained the problem of
astigmatism. It was an eye defect which Airy suffered from himself and he had been the
first to design glasses to correct it. He had earlier published a paper On a peculiar
Defect in the Eye on this problem for which he was the first to provide a practical
solution.
In 1828 Peacock informed Airy that Woodhouse, the Plumian Professor of Astronomy,
had not long to live and advised him to seek this chair. He wrote [3]:-
I made it known that I was a candidate and nobody thought it worthwhile to oppose me.
... I told everyone that the salary (about 300) was not sufficient and drafted a manifesto
to the University for an increase. ... the University had never before been taken by storm
in such a manner and there was some commotion about it. I believe that very few people
would have taken that step. ... I had no doubt of success.
Airy was appointed Plumian Professor of Astronomy at Cambridge and Director of the
Cambridge Observatory. The University had complied with his request for the salary to
be increased and the 500 per year he received was sufficient to allow him to marry
Richarda Smith , which he did on 24 March 1830.
He became Astronomer Royal in 1835 moving at that time from Cambridge to
Greenwich. There he undertook a reorganisation of the Royal Observatory which was
positive in many ways but also had some unfortunate side effects. Since he could not
tolerate his staff thinking for themselves no young scientists were trained at the
Observatory during his period as Astronomer Royal. However, his considerable
engineering ability was put to good use in renovating the instruments at the observatory.
He held this post of Astronomer Royal until 1881 when he resigned and lived the rest of
his life with his two unmarried daughters in the White House close to Greenwich Park.
Airy wrote the text On the Algebraic and Numerical Theory of Errors of Observations
and the Combinations of Observations. Although said at the time to be:-
... unreadable except by those already thoroughly acquainted with the subject,
the book was used at Cambridge and influenced Pearson. This text was one of eleven
books which Airy published, some of the others being Trigonometry (1825), Gravitation
(1834), and Partial differential equations (1866). His remarkable publication record
included over 500 papers and reports. This resulted from his extremely hard work and
also his highly organised, efficient way of working which enabled him to get through
far more work than almost every other scientist.
His attitude to mathematics was very much as an applied mathematician who saw no
point in the study of the subject in its own right. His son writes in [3]:-
His nature was eminently practical, and his dislike of mere theoretical problems and
investigations was proportionally great. He was continually at war with some of the
Cambridge mathematicians on this subject. Year after year he criticised the Senate
House papers and the Smith's Prize papers very severely, and conducted an interesting
and acrimonious private correspondence with Professor Cayley on the same subject.
Airy's delay, in 1845, of searching for Neptune at the location suggested by Adams
prevented Adams obtaining full credit for his work although in many ways he has been
unfairly criticised over this episode. Airy did, however, make many major contributions
to mathematics and astronomy. He improved the orbital theory of Venus and the Moon,
studied interference fringes in optics, made a mathematical study of the rainbow and
computed the density of the Earth by swinging a pendulum at the top and bottom of a
deep mine. We should note that the value he obtained was too large by a fair amount.
Airy was made chairman of the Commission set up to construct Standard Weights and
Measures in 1834. He was elected a Fellow of the Royal Society of Edinburgh in 1835,
 CAPÍTULO VI - 32

and a Fellow of the Royal Society of London in 1836, having received the Society's
Copley Medal in 1831. He gave the Bakerian lecture to the Society entitled On the
theoretical explanation of an apparent new polarity of light in 1840. He received the
Society's Royal Medal in 1845 for a paper on the Irish tides.
The Royal Astronomical Society elected Airy to be their President in 1845. Then, in
1851, Airy was elected President of the British Association, and in 1871 he was elected
President of the Royal Society of London holding the post for two years. The Institut de
France elected him to membership to fill the position which became vacant on the death
of John Herschel in 1872 and in the same year he accepted a knighthood having
declined it on three previous occasions on the grounds that he could not afford the fees.
Soon after this, in 1874, he organised an expedition to observe the transit of Venus.
Outside his professional scientific interests, Airy was a man of broad tastes. He liked
poetry, history, theology, antiquities, architecture, engineering, and geology. He even
published papers on his other interests including one which tried to identify the exact
place where Julius Caesar landed in Britain and the exact place from which he left. In
addition he published a number of papers on religious matters.
There were certainly sides to his character which made him unpopular with those
around him. We have already mentioned how he was a snob at school. In later life he
was sarcastic and enforced a rigid discipline on his staff at the Royal Observatory. In his
defence we would have to note that he enforced such a rigid discipline on himself that it
must have seemed natural to him to expect the same from others.
An illustration of Airy's personality is shown from his long running disagreements with
Babbage. They had a dispute over the quality of a telescope in 1832, he stated that
Babbage's calculating engine was "worthless" ten years later and effectively stopped
government funding of the project, and in 1854 he took the side of the narrow gauge for
railways while Babbage supported the wide gauge. In all these disputes Airy came out
the winner, but it is far from clear that he took the "right" side in the arguments.
We should end with a few words on Airy's importance as a scientist. His own words
certainly show that he had a realistic view of himself (see for example [1]):-
... in those parts of astronomy which ... [require] only method and judgement, with very
little science in the persons employed, we have done much; while in those which
depend exclusively on individual effort we have done little ... our principal progress has
been made in the lower branches of astronomy while to the higher branches of science
we have not added anything.
Eggen writes in [1]:-
Airy was not a great scientist, but he made great science possible.
However, others have a higher opinion of Airy's achievements. Chapman [6] believes
that:-
Airy has been unfairly relegated to the scientific sidelines ...
His son summed up Airy's life as follows [3]:-
The life of Airy was essentially that of a hard-working business man, and differed from
that of other hard-working people only in the quality and variety of his work. It was not
an exciting life, but it was full of interest ...
 CAPÍTULO VI - 33

Richard von Mises

Born: 19 April 1883 in Lemberg, Austria (now Lvov, Ukraine)
Died: 14 July 1953 in Boston, Massachusetts, USA

Richard von Mises was born in Lvov when it was under Austrian control and known as
Lemberg. His father, Arthur Edler von Mises, worked for the Austrian State Railways
as a technical expert and his mother was Adele von Landau. Richard was the second son
of the family, the elder son being Ludwig von Mises who went on to become as famous
as Richard. Ludwig, who was about eighteen months older than Richard, went on to
become an economist who contributed to liberalism in economic theory and made his
belief in consumer power an important part of that theory. Richard also had a younger
brother, who died as an infant.
It was on the technical course that von Mises embarked, studying mathematics, physics
and engineering at the Technische Hochschule in Vienna. After graduating he was
appointed as Georg Hamel's assistant in Brünn. The city of Brünn is today called Brno
in the south-eastern Czech Republic. Up to World War II the inhabitants were
predominantly German, although today they are now mainly Czech. Von Mises was
awarded a doctorate from Vienna in 1907 and the following year he was awarded his
habilitation from Brünn, becoming qualified to lecture on engineering and machine
construction.
He was professor of applied mathematics at Strasburg from 1909 until 1918, although
this was a period which was interrupted by World War I. Even before the outbreak of
the war von Mises had qualified as a pilot and he gave the first university course on
powered flight in 1913. When war broke out von Mises joined the Austro-Hungarian
army and piloted aircraft. He had lectured on the design of aircraft before the war and
he now put this into practice leading a team which constructed a 600-horsepower plane
for the Austrian army in 1915.
After the end of the war von Mises was appointed to a new chair of hydrodynamics and
aerodynamics at the Technische Hochschule in Dresden. Appointed in 1919 he soon
moved again, this time to the University of Berlin to become the director of the new
Institute of Applied Mathematics which had been set up there. Schmidt had argued for
the setting up of the Institute in 1918:-
The pervasion of practical life by mathematical methods, as a result of the development
of technology before the war and, above all, the unexpected need for ... mathematicians
during the war make it an undeniable necessity to install applied mathematics at the
largest Prussian university... Among university students, however, one frequently finds
the opinion that applied mathematics is a subject of inferior importance, which does not
 CAPÍTULO VI - 34

require one's full attention. to create a new tradition, it needs an important personality of
approved name. Such a personality can only be attracted by a full professorship.
Theses words by Schmidt show great wisdom, and the authorities did indeed create the
full professorship and made an inspired choice for the first holder in von Mises.
Ostrowski wrote in a 1965 lecture (see for example [16]):-
Only with the appointment of Richard von Mises to the University of Berlin did the first
serious German school of applied mathematics with a broad sphere of influence come
into existence. Von Mises was an incredibly dynamic person and at the same time
amazingly versatile like Runge. He was especially well versed in the realm of
technology.
The Institute of Applied Mathematics flourished under his control. In 1921 he founded
the journal Zeitschrift für Angewandte Mathematik und Mechanik and he became the
editor of the journal. In the first edition he wrote an introduction stressing the wide
range of applied mathematics and also the fact that the line between pure and applied
mathematics is not a fixed one, but one which changes over time as different areas of
"pure mathematics" find applications in practical situations.
He set up a new curriculum for applied mathematics at the university which spread over
six semesters and included applications of mathematics to astronomy, geodesy and
technology. It was not a "soft option" and von Mises went out of his way to stress that
applied mathematics was every bit as rigorous as pure mathematics requiring [13]:-
... a mathematical model of the widest possible generality, where the argument could be
made with clarity, elegance, and rigour.
His Institute rapidly became a centre for research into areas such as probability,
statistics, numerical solutions of differential equations, elasticity and aerodynamics.
Von Mises was also an excellent lecturer. Collatz, one of his students, wrote:-
I was enrolled in Berlin in 1930. ... Professor Dr Richard von Mises [gave] excellent,
very clear and stimulating lectures on applied analysis ....
The paper [8], written by Collatz, discusses von Mises' work on numerical mathematics,
discusses his founding of the journal Zeitschrift für Angewandte Mathematik und
Mechanik and looks at the difficulties faced by von Mises in bringing up the status of
applied mathematics during the 1920s and early 1930s.
On 30 January 1933, however, Hitler came to power and on 7 April 1933 the Civil
Service Law provided the means of removing Jewish teachers from the universities, and
of course also to remove those of Jewish descent from other roles. All civil servants
who were not of Aryan descent (having one grandparent of the Jewish religion made
someone non-Aryan) were to be retired. Von Mises in one sense was not Jewish for he
was a Roman Catholic by religion. He still fell under the non-Aryan definition of the act
but there was an exemption clause which exempted non-Aryans who had fought in
World War I. Von Mises certainly qualified under this clause and it would have allowed
him to keep his chair in Berlin in 1933. He realised, quite correctly, that the exemption
clause would not save him for long. On the 10 June 1933 he wrote to von Kármán about
a young German, Walter Tollmien, who was looking for a position:-
I have to advise you that the irrevocable prerequisite for any kind of employment or
scholarship or suchlike is to make a statement on his word of honour that his four
grandparents are Aryan and in particular are of non-Jewish descent. ... I believe that in a
favourable case the prospects are not quite so bad as indeed a large part of all the
previous candidates can be omitted under the present law.
Von Kármán forwarded the letter to Tollmien, writing "Indeed a document of our time"
on the back!
 CAPÍTULO VI - 35

Von Mises saw an offer of a chair in Turkey as a way out of his predicament in
Germany but he tried to ensure that his pension rights were preserved. On 12 October
1933 he wrote to the ministry explaining that it would benefit Germany if he accepted a
post in Turkey and that he should be allowed to retain his pension rights for his 24 years
of service. He received the reply that he would have to relinquish all rights of a salary, a
pension or support for his dependants. He protested in a further letter to the Ministry
that he was legally entitled to rights that he was not prepared to relinquish. The Nazi
Theodor Vahlen wanted to take over as director of the Institute despite poor academic
qualities. He promised von Mises that if he would support him to succeed as Director of
the Institute then he would ensure that von Mises would not lose his pension rights.
In October 1933 von Mises wrote his letter to support Vahlen as his successor. Collatz,
von Mises' student, wrote:-
I took my Staatsexamen in November 1933, and Professor von Mises examined me on
the day before his departure. The same day, he talked to me for about one hour, giving
advice for my further research ...
Vahlen was appointed Director of the Institute in December 1933. Having taken up the
new chair in Istanbul, von Mises received a letter in January 1934 denying him any
rights at all. It was something that von Mises continued to feel extremely aggrieved
about, writing to the Ministry in 1953, shortly before his death, still trying to restore his
rights.
The mathematician Hilda Geiringer followed him to Istanbul in 1934. There she was
appointed as professor of mathematics.
In 1938 Kemal Atatürk died and those in Turkey who had fled from the Nazis feared
that their safe haven would become unsafe. In 1939 von Mises left Turkey for the
United States. He became professor at Harvard University and in 1944 he was
appointed Gordon-McKay Professor of Aerodynamics and Applied Mathematics there.
Geiringer followed him to the USA and they were married in 1943.
Von Mises worked on fluid mechanics, aerodynamics, aeronautics, statistics and
probability theory. He classified his own work, not long before his death, into eight
areas: practical analysis, integral and differential equations, mechanics, hydrodynamics
and aerodynamics, constructive geometry, probability calculus, statistics and
philosophy. He introduced a stress tensor which was used in the study of the strength of
materials. His studies of wing theory for aircraft led him to investigate turbulence.
Much of his work involved numerical methods and this led him to develop new
techniques in numerical analysis. His most famous, and at the same time most
controversial, work was in probability theory.
He made considerable progress in the area of frequency analysis which was started by
Venn. He combined the idea of a Venn limit and a random sequence of events.
Ostrowski in the same lecture which we quoted from above wrote (see for example
[16]):-
Because of his dynamic personality his occasional major blunders were somehow
tolerated. One has even forgiven him his theory of probability.
This judgement by Ostrowski is rather harsh, however, and many consider von Mises'
theories to be important in the development of the subject. After the measure theory
approach by Kolmogorov had become favoured by almost all statisticians over von
Mises' limiting frequency theory approach, there was a return to von Mises ideas and
there was an attempt to incorporate them into the measure theoretic approach of
Kolmogorov who wrote himself in 1963:-
... that the basis for the applicability of the results of the mathematical theory of
probability to real 'random phenomena' must depend on some form of the frequency
 CAPÍTULO VI - 36

concept of probability, the unavoidable nature of which has been established by von
Mises in a spirited manner.
In paper [18] discusses:-
... von Mises' notion of a random sequence in the context of his approach to probability
theory. [The author claims] that the acceptance of Kolmogorov's rival axiomatisation
was due to a different intuition about probability getting the upper hand, as illustrated
by the notion of a martingale.
Phillip Frank, writing in [4] says:-
... in looking over the work of von Mises ... we cannot fail to recognise a whole
spectrum of research, extending from the philosophical meaning of science to practical
methods of numerical computation. ... von Mises has always been a truly broad-minded
man ... notwithstanding the wide range of his topics, his work shows a great intrinsic
unity: starting from a definite center, it branches out in systematic investigations of a
great diversity of problems. ... von Mises chose the topics according to a very definite
view-point, determined by his ideas about the essence and method of every thoroughly
scientific research.
In von Mises' book Positivism: A Study in Human Understanding (1951) he expressed
his views on science and life. He subscribes to a doctrine of positivism in the book
saying:-
Positivism does not claim that all questions can be answered rationally, just as medicine
is not based on the premise that all diseases are curable, or physics does not start out
with the postulate that all phenomena are explicable. But the mere possibility that there
may be no answers to some questions is no sufficient reason for not looking for
answers, or for not using those which are attainable.
This interest in philosophy was only one of von Mises' interests outside the realm of
mathematics. Another was the fact that he was an international authority on the Austrian
poet Rainer Maria Rilke (1875-1926).
In 1950 von Mises was offered honorary membership of the East German Academy of
Science. This was difficult for von Mises, particularly in McCarthy era America where
any link with communism would have been viewed as a serious offence. He sadly
declined in a letter written on 15 September 1950:-
I would very gladly accept the nomination in remembrance of my teaching activities in
Berlin and thus re-establish the bond which connected me for a long time to the German
scientific life. Unfortunately the present circumstances in Germany as well as those in
this country are such that the acceptance of such a distinction could be interpreted as a
political demonstration on my part. ... I only relinquish acceptance of this nomination
under the pressure of outward circumstances, a nomination which I regard as a great
honour in every respect.
 CAPÍTULO VI - 37

Jean Victor Poncelet

Born: 1 July 1788 in Metz, Lorraine, France
Died: 22 Dec 1867 in Paris, France

Jean-Victor Poncelet was a pupil of Monge. His development of the pole and polar lines
associated with conics led to the principle of duality.
Poncelet took part in Napoleon's 1812 Russian campaign as an engineer. He was left for
dead at Krasnoy and imprisoned until 1814 when he returned to France. During his
imprisonment he studied projective geometry. He also wrote a treatise on analytic
geometry Applications d'analyse et de géométrie based on what he had learnt at the
École Polytechnique but it was only published 50 years later.
From 1815 to 1825 he was a military engineer at Metz and from 1825 to 1835 professor
of mechanics there. He applied mechanics to improve turbines and waterwheels more
than doubling the efficiency of the waterwheel.
Poncelet was one of the founders of modern projective geometry simultaneously
discovered by Joseph Gergonne and Poncelet. His development of the pole and polar
lines associated with conics led to the principle of duality. He also discovered circular
points at infinity.
He published Traité des propriétés projectives des figures in 1822 which is a study of
those properties which remain invariant under projection. This work contains
fundamental ideas of projective geometry such as the cross-ratio, perspective, involution
and the circular points at infinity. While writing this book he consulted with Servois.
Poncelet published Applications d'analyse et de géométrie in two volumes: 1862 and
1864.
 CAPÍTULO VI - 38

George Gabriel Stokes

Born: 13 Aug 1819 in Skreen, County Sligo, Ireland

Died: 1 Feb 1903 in Cambridge, Cambridgeshire, England

George Stokes' father, Gabriel Stokes, was the Protestant minister of the parish of
Skreen in County Sligo. His mother, Elizabeth Haughton, was the daughter of a minister
of the church, so George Stokes' upbringing was a very religious one. He was the
youngest of six children and every one of his three older brothers went on to become a
priest. As the priest of the church in Cambridge which Stokes later attended wrote (see
[15]):-
Though he was never narrow in his faith and religious sympathies, he always held fast
by the simple evangelical truths he learnt from his father...
In [4] the atmosphere in which George grew up is described in words which are more
colourful than those which might be used today:-
The home-life in the Rectory at Skreen was very happy, and the children grew up in the
fresh sea-air with well-knit frames and active minds. Great economy was required to
meet the educational needs of the large family...
It was not only religious teaching, but a wider introduction to education, which Gabriel
Stokes was able to give his children. In particular, having studied at Trinity College
Dublin, he was able to teach George Latin grammar. Before going to school George was
also taught by the clerk in his father's parish in Skreen. Leaving Skreen in 1832, George
attended school in Dublin. He spent three years at the Rev R H Wall's school in Hume
Street, Dublin, but he was not a boarder at the school, living for these three years with
his uncle John Stokes. In fact the family finances would not have allowed him a more
expensive education, but at this school [4]:-
He pursued the usual school studies, and attracted the attention of the mathematical
master by his solution of geometrical problems.
It was during George's three years in Dublin that his father died and this event had, as
one would expect, a major effect on the young man.
In 1835, at the age of 16, George Stokes moved to England and entered Bristol College
in Bristol. The two years which Stokes spent in Bristol at this College were important
ones in preparing him for his studies at Cambridge. The Principal of the College, Dr
Jerrard, was an Irishman who had attended Cambridge University with William Stokes,
one of George's elder brothers. Dr Jerrard was himself a mathematician but Stokes was
taught mathematics at Bristol College by Francis Newman (who was the brother of John
 CAPÍTULO VI - 39

Henry Newman, later Cardinal Newman, who became the leader of the Oxford
Movement in the Church of England which was founded in 1833). Clearly Stokes talent
for mathematics was shown during his studies at Bristol College, for he won
mathematics prizes and Dr Jerrard wrote to him (see [4]):-
I have strongly advised your brother to enter you at Trinity, as I feel convinced that you
will in all human probability succeed in obtaining a Fellowship at that College.
It was not Trinity, rather Pembroke College, Cambridge, which Stokes entered in 1837.
There are slight inconsistencies in what his mathematical background was on entering
Cambridge. In the course at Bristol College (according to the College literature) (see
[5]):-
... a student was to become acquainted with the differential and integral calculus and to
go on to statics, dynamics, conic sections and the first three sections of Newton's
Principia...
However, Stokes himself wrote in 1901 (see for example [4]):-
I entered Pembroke College, Cambridge in 1837. In those days boys coming to the
University had not in general read so far in mathematics as is the custom at present; and
I had not begun the differential calculus when I entered the College, and had only
recently read analytical sections.
In Stokes' second year at Cambridge he began to be coached by William Hopkins, a
famous Cambridge coach who played a more important role than the lecturers. Stokes
wrote [4]:-
In my second year I began to read with a private tutor, Mr Hopkins, who was celebrated
for the very large number of his pupils gaining high places in the University
examinations for mathematical honours...
Hopkins was to exert a strong influence on the direction of Stokes' mathematical
interests. Hopkins [5]:-
... praised the study of physical astronomy and physical optics, for example, because
they revealed mathematics to be 'the only instrument of investigation by which man
could possibly have attained to a knowledge of so much of what is perfect and beautiful
in the structure of the material universe, and the laws that govern it'.
In 1841 Stokes graduated as Senior Wrangler (the top First Class degree) in the
Mathematical Tripos and he was the first Smith's prizeman. Pembroke College
immediately gave him a Fellowship. He wrote [4]:-
After taking my degree I continued to reside in College and took private pupils. I
thought I would try my hand at original research....
It was William Hopkins who advised Stokes to undertake research into hydrodynamics
and indeed this was the area in which Stokes began to work. In addition to Hopkins'
advice, Stokes was also inspired to enter this field by the recent work of George Green.
Stokes published papers on the motion of incompressible fluids in 1842 and 1843, in
particular On the steady motion of incompressible fluids in 1842. After completing this
research Stokes discovered that Duhamel had already obtained similar results but, since
Duhamel had been working on the distribution of heat in solids, Stokes decided that his
results were obtained in a sufficiently different situation to justify him publishing.
Stokes then continued his investigations, looking at the situation where he took into
account internal friction in fluids in motion. After he had deduced the correct equations
of motion Stokes discovered that again he was not the first to obtain the equations since
Navier, Poisson and Saint-Venant had already considered the problem. In fact this
duplication of results was not entirely an accident, but was rather brought about by the
lack of knowledge of the work of continental mathematicians at Cambridge at that time.
Again Stokes decided that his results were obtained with sufficiently different
 CAPÍTULO VI - 40

assumptions to justify publication and he published On the theories of the internal

friction of fluids in motion in 1845. The work also discussed the equilibrium and motion
of elastic solids and Stokes used a continuity argument to justify the same equation of
motion for elastic solids as for viscous fluids.
Perhaps the most important event in the recognition of Stokes as a leading
mathematician was his Report on recent researches in hydrodynamics presented to the
British Association for the Advancement of Science in 1846. But a study of fluids was
certainly not the only area in which he was making major contributions at this time. In
1845 Stokes had published an important work on the aberration of light, the first of a
number of important works on this topic. He also used his work on the motion of
pendulums in fluids to consider the variation of gravity at different points on the earth,
publishing a work on geodesy of major importance On the variation of gravity at the
surface of the earth in 1849.
In 1849 Stokes was appointed Lucasian Professor of Mathematics at Cambridge. In
1851 Stokes was elected to the Royal Society, awarded the Rumford medal of that
Society in 1852, and he was appointed secretary of the Society in 1854. The Lucasian
chair paid very poorly so Stokes needed to earn additional money and he did this by
accepting an additional position to the Lucasian chair, namely that of Professor of
Physics at the Government School of Mines in London.
Stokes' work on the motion of pendulums in fluids led to a fundamental paper on
hydrodynamics in 1851 when he published his law of viscosity, describing the velocity
of a small sphere through a viscous fluid. He published several important investigations
concerning the wave theory of light, such as a paper on diffraction in 1849. This paper
is discussed in detail in [11] in which the authors write:-
...the results of Stokes are related to the elastic theory of light, and supplement and
expand a number of questions, previously studied for the most part in the works of A
Cauchy. Stokes's methods for solving diffraction problems, differing considerably from
the methods employed by Cauchy, form the basis of the further studies of the
mathematical theory of the phenomenon of diffraction.
Stokes named and explained the phenomenon of fluorescence in 1852. His
interpretation of this phenomenon, which results from absorption of ultraviolet light and
emission of blue light, is based on an elastic aether which vibrates as a consequence of
the illuminated molecules. The paper [9] discusses this in detail and is particularly
interesting since the author makes full use of Stokes' unpublished notebooks.
In 1854 Stokes theorised an explanation of the Fraunhofer lines in the solar spectrum.
He suggested these were caused by atoms in the outer layers of the Sun absorbing
certain wavelengths. However when Kirchhoff later published this explanation, Stokes
disclaimed any prior discovery.
Stokes' career certainly took a rather different tack in 1857 when he moved from his
highly active theoretical research period into one where he became more involved with
administration and experimental work. Certainly his marriage in 1857 was not
unconnected with the change in tack and, particularly since it gives us an insight into
Stokes' personality, we shall look at the events. Stokes became engaged to marry Mary
Susanna Robinson, the daughter of the astronomer at Armagh Observatory in Ireland. In
[4] a number of letters from Stokes to Mary Susanna Robinson are given. On 21
January 1857 he wrote of his feelings for her:-
I was capable of being moved, mathematically, as it were, by the belief that a particular
course was right; and I do believe that God put these views in my mind, working by
means of that which was in me to supply that which was wanting.
Three days later he wrote that she had stopped him becoming an old bachelor:-
 CAPÍTULO VI - 41

I feel that perhaps my marriage with you would be even the turning-point of my
salvation.
A further three days later he wrote:-
You are quite right in saying that it is well not to go brooding over one's own thoughts
and feelings, and in a family that is easy, but you don't know what it is to live utterly
alone.
On the 31 March 1857 he wrote again expressing his feelings in rather mathematical
terms:-
I too feel that I have been thinking too much of late, but in a different way, my head
running on divergent series, the discontinuity of arbitrary constants, ... I often thought
that you would do me good by keeping me from being too engrossed by those things.
These letters clearly did not express the love that Mary hoped to find in them and when
Stokes wrote her a 55 page letter (which was possibly deliberately destroyed) about the
duty he felt towards her, she came close to calling off the wedding at the last moment.
On receiving her letter showing that she was unhappy to go ahead with the marriage
Stokes replied:-
Then it is right that you should even now draw back, nor heed though I should go to the
grave a thinking machine unenlivened and uncheered and unwarmed by the happiness
of domestic affection.
The marriage did go ahead and Stokes certainly turned away from his life of intense
mathematical research. It may appear from the above quotations that in fact Stokes was
really looking for this change in his life and perhaps he sought marriage partly so that
this change in his life-style could come about.
Now Fellows at Cambridge had to be unmarried, and so on his marriage in 1857 Stokes
had to give up his fellowship at Pembroke College. However, a change in the rules in
1862 allowed married men to hold fellowships and he was able to take up the fellowship
at Pembroke again. Stokes continued as secretary of the Royal Society from his
appointment in 1854 until 1885 when he was elected President of the Society. He held
the position of President until 1890. He was also president of the Victoria Institute from
1886 until his death in 1903. There were other administrative tasks which he undertook.
In 1859 he had written to Thomson saying:-
I have another iron in the fire now: I have just been appointed an additional secretary of
the Cambridge University Commission.
P G Tait mentioned this in his criticism of the way that science was organised in Britain
[16]:-
What a comment on things as they are is furnished by the spectacle of genius like that of
Stokes' wasted on the drudgery of Secretary to the Commissioners for the University of
Cambridge; or of a Lecturer in the School of Mines; or the exhausting labour and totally
inadequate remuneration of a Secretary to the Royal Society.
Stokes received the Copley medal from the Royal Society of London in 1893 and he
was given the highest possible honour by his College when he served as Master of
Pembroke College in 1902-3.
Stokes' influence is summed up well by Parkinson in [1]:-
... Stokes was a very important formative influence on subsequent generations of
Cambridge men, including Maxwell. With Green, who in turn had influenced him,
Stokes followed the work of the French, especially Lagrange, Laplace, Fourier, Poisson,
and Cauchy. This is seen most clearly in his theoretical studies in optics and
hydrodynamics; but it should also be noted that Stokes, even as an undergraduate,
experimented incessantly. Yet his interests and investigations extended beyond physics,
 CAPÍTULO VI - 42

for his knowledge of chemistry and botany was extensive, and often his work in optics
drew him into those fields.
One notable omission from his publication list was a treatise on light. This omission
was in part due to the change in his research output after 1857 but it was also partly due
to not wishing to report upon speculative ideas in a field which was in a rapid state of
progress. Stokes' failure to publish a treatise on optics is discussed in detail in [7].
However, he did lecture on optics in his Burnett lectures at the University of Aberdeen
in 1891-93 and these lectures were published.
Stokes' mathematical and physical papers were published in five volumes, the first three
of which Stokes edited himself in 1880, 1883 and 1891. The last two were edited by Sir
Joseph Larmor with the work being completed in 1905.
These comments about Stokes' character in [3] are interesting:-
From 1887 to 1892 he was one of the members of Parliament for Cambridge University.
In spite, or perhaps because of, his great and profound knowledge and remarkable
ability, he rarely spoke in the House of Commons, but was always listened to with
attention. In private life his simplicity and modesty were as conspicuous as his great
attainments.