532 



NATURE 



[December 23, 1920 



Letters to the Editor. 



{The Editor does not hold himsel/ responsible for opinions 

 expressed by his correspondents. Neither can he undertake to 

 renim, or to correspond with the writers ol, rejected manu- 

 scripts Intended /or this or any other part ol NATURE. No 

 notice is taken of anonymous communications. \ 



Heredity and Acquired Characters. 



Although agreeing enurtly with Sir Ray Lan- 

 kester's and i'rof. MacBride's criticisms (Natlkk, 

 December i6, pp. 500-1) of Sir Archdall Reid's letter, 

 I venture to add a few words on two points. 



First, 1 will attempt to explain to Sir Archdall Reid 

 why it is held that a head is "more innate and ger- 

 minal and less acquired and somatic than a scar " 

 (ibid., November 25, p. 405). 



It is reasonable to suppose that, at least up to 

 hatching, the nurture of the chicks of all domestic 

 races is the same. The differences between their 

 heads af-e therefore due, not to nurture, but to nature. 

 Nurture is an essential condition of their develop- 

 ment, but, being uniform, it cannot be a cause of 

 diversity. 



.\ scar is the precise opposite of this. Its cause is 

 the blow or other e.xternal force inflicting injury, an 

 essential condition being the existence of organic 

 matter incapable of resisting the force and with 

 limited powers of regeneration. This condition is 

 uniform over a large part of the body surface, and 

 can never explain the infinite variety in the appear- 

 ance and position of scars. 



We therefore rightly maintain that everything dis- 

 tinctive of the normal head, everything which we can 

 describe, measure, or figure, is inherent or blasto. 

 genie, although appropriate external conditions aie 

 necessary for its development ; and that, similarly, 

 every feature by \Yhich one scar is distinguished from 

 another is acquired or somatogenic, although germinal 

 potentialities are necessary for its existence. 



One other consideration arises out of Sir Ray 

 Lankester's letter. Although I entirely agree that 

 acquired characters are commonly "departures (either 

 increase or decrease) from the usual or normal size, 

 form, or structure of this or that part," vet — as 1 

 pointed out in reply to similar definitions by Sir Francis 

 Galton and Prof. Lloyd Morgan — it is "a mistake to 

 make too much of abnormalitv, or to import it into a 

 definition. Some of the most marked, and certainlv 

 the most easilv studied and tested, of acquired 

 characters are the differences between the effects of 

 alternative environments, all of which are normal, 

 upon the individuals of a single species. The green 

 colour of a larva of Amphidasys betularia, if fed upon 

 broom, is an acquired character, as is the dark colour 

 it would assume upon oak. etc. I think, therefore, 

 that a more satisfactory definition of, at anv rate, a 

 large class of acquired characters mav be framed as 

 follows : ' Whenever change in the environment 

 regularly produces appreciable change in an organism, 

 siirh difference mav be called an acauired character'" 

 (Proc. Ent. Soc. Lond., 1904, p. cviii.). 



Edward B. Poulton. 



Oxford, December ig. 



factors at work, however, are more deep-seated than 

 they appear to be at first sight. That this is so is at 

 once clear from two correlated considerations : 

 (i) There is a. period, sometimes a long period, of 

 ripening with reference to which spawning is the 

 culmination, and during which somatic changes and 

 migratory results are produced ; and (2) all the adults 

 do not participate in the ripening and spawning. 



In the case of the Protista and the Metazoa which 

 reproduce rapidly, an annual phase may be postulated 

 which closely follows the rise and fall of temperature, 

 but in each generation reproduction follows growth, 

 and does not take place until the growth of the cell 

 or of the soma has reached a certain condition. 

 When the growth is represented by a curve the point 

 of inflection may be said to mark the period when, 

 in the case of the cell, the impulse is conveyed to 

 prepare for division, and in the case of the Metazoor\ 

 the call arrives to the gonad to develop. Now 

 amongst the Metazoa we get all stages, from the 

 rapid reproduction of such forms to those which repro- 

 duce twice a year or once a year, and to those which 

 reach maturity after several years of growth and 

 do not necessarily thereafter spawn annually. In 

 each case we have to observe that the somatic need 

 must be satisfied before maturation is introduced. 

 j There are aquatic and terrestrial species which in 

 ! one situation reproduce once a year, and in warm 

 I seasons or in warmer situations reproduce twice a 

 i year. But we understand by that statement that the 

 conditions "have allowed the early generation to grow 



Environment and Reproduction. 



Several attempts have been made to show that 

 spawning in the sea and in fresh waters, and also in 

 terrestrial conditions, is regTjlated by temperature. 

 Many examples could be quoted to demonstrate that 

 the rule is for a species to spawn earlier and to 

 develop more quickly in warm water ; that, in other 

 words, the period of spawning is early in tropical 

 seas and conditions, and later to the north and south 

 of the tropical belt, and that the period of growth 

 is increased with lowering of temperature. The 

 NO. 2669, VOL. 106] 



sufficiently fast to permit of a second generation in 

 the same season. 



In the accompanying diagrams (Fig. 1) an attempt 

 is made to indicate the effect of summer and winter 

 on Protozoa (A) and on Metazoa which have many 

 generations a year (B). Lowering of temperature in 

 these cases would restrict growth, and therefore the 

 number of generatfons. The diagrams also depict the 

 acceleration and retardation of growth and reproduc- 

 tion of planktonic species passing to and from tropical 

 seas. In (B) the history of the gonad is indicated by 

 the black curves. This is also done in (C), which is 

 meant to show the periodical changes in a species 

 which spawns regularly. 



With regard to the last, we must first differentiate 

 between the individual and the crowd. The spawning 

 period may be a long one in the tropics, but it is 

 periodic for the individual, and, indeed, spawning 

 seasons are in many cases sharply indicated in 

 tropical species of all g'roups. If, with reference to 

 such an example as (C), the season of the commence- 

 ment of maturation of the gonad be approximately 

 constant, the effect of lowering of temperature would 

 be manifested by a later spawning. Such con- 

 siderations go, therefore, a long way to explain 

 the observed results, and they also serve to 

 explain the length of the tropical spawning season 

 and the more restricted season of temperate, and 

 especially of Arctic, regions. Without reference to 



