NATURE 



[June io, 1922 



grip that it persists after the inducing conditions have 

 ceased to operate. It may be advisable to drop the 

 term ' acquired character ' in favour of ' somatic 

 modification/ or the like, but there is not in the mind 

 of the competent biologist any confusion whatsoever 

 in regard to what 'acquired character' means. He 

 uses it as he might use a symbol ' A.C 



It may be of use to try to state in outline the meaning 

 which biologists attach to the terms they use in de- 

 scribing the facts of variation and heredity. When we 

 compare a thousand descendants of two parents of 

 known pedigree (we are evading the difficult question 

 of ' species '), we may find that they are far from being 

 all alike. They may differ from one another and from 

 the norm of the stock to which they belong. These 

 individual peculiarities can be measured and registered, 

 and they might be called ' observed divergences,' 

 ' observed differences,' ' new departures,' and so 

 forth, so as to avoid any question-begging term. They 

 are changes, at any rate, and many biologists apply to 

 them the general word ' variations.' 



When we begin to study the observed differences 

 critically, we find that some of them are directly due 

 to nurtural peculiarities. They are novelties due to 

 novel conditions, — of use and disuse, or of food, or of 

 surroundings. This individual has been imprisoned 

 and that one has been overworked ; this individual has 

 been starved and that one overfed ; this one has been 

 brought up in a cave and that one exposed to the sun. 

 The resulting differences are somatic modifications or 

 acquired characters. They are exogenous, imprinted 

 from without, what Weismann called ' somatogenic,' 

 what might be called dints in colloquial biology. The 

 goldfish kept three years in darkness loses the rods and 

 cones of the retina ; the tropical explorer becomes per- 

 manently tanned ; the peach-trees taken to Reunion 

 became evergreens, though it took some of them twenty 

 years. 



Every character is the product of hereditary 

 ' factors ' and a certain minimum nurture — food, 

 oxygen, water, and a succession of liberating stimuli ; 

 but ' modifications ' are lasting changes wrought on 

 the body of the individual as the direct result of 

 some peculiarity in the conditions of life. A modifica- 

 tion is more than a response to liberating stimuli ; it 

 is an imprint, a dint, a parry to a nurtural thrust. 

 The development of the head will only occur within 

 certain limits — the primary conditions of viability ; 

 but the racial peculiarities of the skull are in a different 

 category from deformations imprinted before or after 

 birth. Sir Archdall Reid says this is a distinction 

 without a difference. But we think he is making fun 

 of us. 



Of course none of these distinctions is absolutely 

 NO. 2745, VOL. 109] 



hard and fast, for we are dealing with the complexity 

 of life. Thus it is not easy to draw a firm line between 

 temporary adjustments like sunburning, and lasting 

 modifications like tanning. On the other side, there 

 are great differences in the degree in which the 

 developmental expression of hereditary ' factors ' 

 demands particular nurtural conditions. Up to a 

 certain point the development of the lung of the 

 embryo chick proceeds without functioning — the 

 gramophone of the inheritance continuing to unwind 

 after the spring has been released. But beyond a 

 certain point the actualisation of the hereditary lung 

 ' factors ' will not go unless the chick breathes. For a 

 short time an embryo rabbit will go on developing out- 

 side the mother ; which shows that the environmental 

 dependence is not so narrowly limited as one might 

 think. On the other hand, there are several well-known 

 mutations in the fruit-fly Drosophila which never find 

 expression unless there is some particular external 

 condition, such as humidity. As the next generation 

 shows, the primordium of the character is part of the 

 inheritance, but it remains latent or unexpressed. 



Now when we add up all the peculiarities that can 

 be reasonably interpreted as modifications and subtract 

 them from the total of observed differences, we get a 

 remainder — variations as opposed to modifications ; 

 or, if the term variations has been already used more 

 widely, germinal variations as opposed to Lamarck's 

 " changements acquis." They occur among individuals 

 living under the same nurtural conditions. They arise 

 from within, endogenously, centrifugally ; they are 

 ' blastogenic ' in Weismann's phraseology ; they are 

 expressions of new permutations and combinations 

 among the germinal ' factors,' ' genes,' or ' deter- 

 minants ' ; they may be expressions of changes in the 

 constitution of a 'gene' itself. We know very little 

 as to the origin of these germinal variations sensu 

 stricto, but there is no confusion in our concept. Then 

 we may proceed further to distinguish quantitative 

 gametogenic fluctuations from qualitative gametogenic 

 mutations, meaning by the former a little more of this 

 and a little less of that (a longer feather, a higher crest), 

 meaning by the latter a discontinuous or transilient 

 variation, such as a laciniate condition in the Greater 

 Celandine leaves or hornlessness of calves in a horned 

 race. But in the present state of ignorance the cate- 

 gories of variation are necessarily very debatable. 



The scheme runs thus :— 



^GAMETOGENIC [FLUCTUATIONS 



OBSERVED J VARIATIONS j MUTATIONS 



DIVERGENCES 1 SOMATIC 



I MODIFICATIONS 



What terms are used does not seem urgently important ; 

 there has often to be a long struggle for existence among 



