October 26, 1905] 



NA TURE 



633 



ensues. ... (8) The variations produced by food in- 

 crease in intensity with each generation, and even 

 arrive at sucii a point as to persist to a degree, by 

 heredity, in the next generation brought up on normal 

 food; when, in successive generations, the food plant 

 is different, each kind of food plant impresses its 

 characteristic effects on the imago. (9) After some 

 generations on the abnormal food the insect becomes 

 accustomed to it, and this brings about a return to 

 the primitive type — sometimes, indeed, passes beyond 

 it in the opposite direction. 



The experiments which led to these conclusions 

 extended over five years, from 1900 to 1904, and were 

 tried on 21 different species and about 4695 in- 

 dividuals. The paper is illustrated bv five plates con- 

 taining eighty-one photographic figures, which are 

 excellent, but uncoloured, so that they have not the 

 advantage of showing the distinctive colour effects 

 which enter into the verbal description of the results 

 obtained. The course of experiment can only be briefly 

 indicated here, having due regard to the exigencies of 

 space, but I may select for reference some of M. 

 Pictet's chief experiments on what was their principal 

 subject, Ocneria dispar ; on this species there were 

 twenty-nine experiments upon 1568 individuals. In 

 many of those tried on this and other species the 

 differences from the normal, so far as they are shown 

 by the plates, are not very distinguishable from those 

 deficiencies in intensity and definiteness of marking 

 and the dwarfing of size that one is accustomed to 

 find when larvse are bred on food that is insufficient 

 or unsuitable, to put it in a popular form, are " half 

 starved." It is right, however, to say that M. Pictet 

 considers, as afterwards mentioned, that in those 

 examples which he has selected for illustration as 

 exhibiting the effects of abnormal food plants, walnut, 

 onobrychis, &c., they are distinguishable from each 

 other to such an extent that where larvae have been 

 fed for three successive generations on walnut, 

 onobrychis, and oak respectively, the special influences 

 of all three food plants can be seen. 



In si.x experiments with O. dispar, walnut was 

 given for one or more generations ; in all these 

 cases the wing expansion was considerably smaller 

 than normal, in some cases not more than 

 three-quarters or two-thirds of it. Where O. sativa, 

 dandelion or P. sanguisorba was given the 

 imagines were considerably larger than normal, 

 but when in one or more of the succeeding 

 generations walnut was substituted the size was 

 immediately reduced, much as in the other six experi- 

 ments. Mespilus germanicus, horse chestnut, white 

 poplar, and sallow had effects very similar to those 

 of walnut. In experiment (4), where oak in the 

 second generation succeeded walnut in the first, there 

 was a slight return towards the type, but when in 

 the third generation walnut was again given, the 

 failure in intensity of markings reached its minimum, 

 there being scarcely a trace of colour; when, how- 

 ever, in the fourth generation oak was again given, 

 there was a nearer return towards the tvpe than the 

 second generation showed. In other cases the 

 " albinistic " influence of the walnut persisted in a 

 very marked degree after two later generations fed 

 on oak or on O. sativa. In such cases, also, where 

 other food plants of the three different classes 

 (" albinising," "normal," and " melanising ") had 

 been given in succession, M. Pictet considers that the 

 special pigmentation effects of each of the three kinds 

 of food plant are shown by the imagines of the latest 

 generation. These are for walnut, c? , pale yellow 

 colour, two central lines partly obliterated, other 

 markings less intense ; 9 , wings slightly transparent, 

 few markings on upper wings, more on lower; second 



NO. 1878, VOL. 72] 



generation, cJ, wings whitish, marginal band on all 

 partly obliterated, transverse lines little visible ; 

 9, wings transparent, the V mark and the marginal 

 dots alone appearing; for O. sativa, d, wings brown, 

 zigzag, lines little noticeable, marginal band very 

 dark, abdominal hairs greyish ; 9 , on upper wings 

 white zigzag lines strongly marked; for dandelion, 

 d , very similar, only the lower wings of uniform 

 dark colour. 



M. Pictet arrives at the general conclusion that the 

 " albinising " variations are caused by the larvae 

 having been fed on leaves presenting obstacles to 

 nutrition, such as hard cuticle or felted underside, as 

 in white poplar, and that, on the other hand, the 

 " melanising " variations are caused by food present- 

 ing no such obstacles ; thus the young leaves of laurel 

 are not " melanising " as the old leaves are. So far 

 as I am aware, M. Pictet's conclusion that a differ- 

 ence of food plant in one generation can cause a 

 difference of facies in the imago, and one that persists 

 for several generations, is not in accordance with 

 views hitherto prevailing; its bearing on the question 

 whether a quality thus "acquired can originate a new 

 permanent variety or species is, however, at least 

 materially affected by M. Pictet's other position, that 

 where several generations have been brought up on 

 the abnormal food so as to become accustomed to it, 

 they revert towards the original form, so that there 

 would appear to be only a temporary disturbance in 

 the colouring of the species. 



All M. Pictet's figures of O. dispar are reproduced 

 as illustrative of this notice; those numbered 13 

 (walnut, oak, onobrychis), 14, and 16 (onobrychis, 

 mespilus) are relied on by him as showing indications 

 of each of the different food plants supplied to them 

 and their ancestors, that numbered 10 (walnut, oak, 

 walnut, walnut) as showing reversion towards the 

 original normal form when the larvae have for several 

 generations been confined to abnormal food. 



With respect to M. Pictet's position that an inverse 

 rate of development in the pupa is caused by lengthen- 

 ing or shortening the duration of the larval " dia- 

 pause " or period of repose, his experiments favour 

 that view; but it will hardly be accepted as of general 

 application without further experiments. 



There is a section on the influence of food on 

 the colour of the larvae in which M. Pictet states 

 that such an influence is exerted, with observ- 

 ations tending to show that in some cases there 

 is a relation between the colour thus induced in 

 the larva and the colouring of the imago. There 

 are also experiments from which he draws the 

 conclusion that the kind of food influences the 

 secondary sexual characters of the larvae which are so 

 marked in O. antiqua, &c. ; this does not, of course, 

 mean that it changes the sex as has been asserted; 

 on that he makes the just observation that it is not 

 sufficient to count the respective numbers of males 

 and females among the perfect insects obtained, but 

 account ought also to be taken of those that die, 

 usually in large numbers, and the male sex may_ be 

 much more capable than the female of supporting 

 the " tribulations of life," among which, one may 

 add, must certainly be included scientific experiments 

 on their food. 



The second part of M. Pictet's paper is devotedto 

 the influence of humidity. Excessive moisture applied 

 to young larvas is largely fatal, but seems to have no 

 effect on the perfect insects which survive, beyond 

 slightly reducing their size. Older larvae, i.e. (usually) 

 for the period of eight or ten days before pupation, 

 resist it perfectly, but give "aberrations," some of 

 which are figured, such as are met with here and 

 there in nature. 



