CATERPILLAR 



day ; others feign death when caught. In more 

 than one species (Dicranura) the larva spurts out 

 formic acid. 



Darwin had leen impressed with the colours of 

 caterpillars, which, occurring as 

 they do on larva 1 , could not he 

 referred to the action of sexual 

 selection. Though, as an excep- 

 tion, a male caterpillar may be 

 much brighter than the female, 

 the two sexes are all but invari- 

 ably the same, and no sexuality 

 or sexual choice is yet devel- 

 oped. Wallace attacked the 

 problem, and maintained that 

 the conspicuous forms were dis- 

 tasteful to birds and reptiles, 

 and that the conspicuous colours 

 were advantageous reminders of 

 unpalatableness. To this Poul- 

 ton has added the necessary 

 caution, that animals forced by 

 hunger will eat the distasteful 

 beauties, and come to like them. 

 For a most interesting series 

 of studies on the colour and 



Fig. 4. Hawthorn 

 twig with at- 

 tached twig-like 

 larva (a) of 

 Jin ni in Cradce- 



yata ( after Poul- markings of caterpillars, the 

 ton). reader should consult Pro- 



fessor Weismann's Studies in 

 the Theory of Descent, and the valuable editorial 

 notes of the translator, Professor Meldola. The 

 whole burden of Weismann's work is to show that 

 in the marking and colouring (of the Sphingidse 

 in particular) no action of an impelling vital force 

 can be recognised, but that ' the origination and 

 perfection of these characters depend entirely on 

 the known factors of natural selection and correla- 

 tion,' though of course natural selection can only 

 operate on the variations possible to the physical 

 constitution and conditions of the organism. In 

 tracing the presumed historical evolution of the 

 Sphingidte, which is more or less fully recapitu- 

 lated in the individual development, he starts ( 1 ) 

 from concealed or subterranean, white or yellow, 

 forms, with a horn on the tail and with bristles, but 

 without markings ; ( 2 ) in adaptation to life on 

 linear plants like grasses, longitudinal markings 

 -are evolved and confirmed by natural selection ; 



Fig. 5. 



o, Caterpillar of Sphinx Convohitii ; b, larva of Macroglossa 

 Stellatarum, showing lines and spots (after Weismann). 



<3) these are succeeded by oblique stripes, spread- 

 ing from one segment to another, evolved by 

 natural selection and correlation, and followed by 

 the disappearance of the longitudinal lines which 

 spoil the effect; (4) on the second last segment 

 ring-spots then appear, and tend to spread to other 

 rings ; these are deceptively like the oerries of the 

 food-plant at one time, or have a terrifying eye- 



like appearance afterwards, and are often aignaU 

 of distiustefulneHH. 



Poul ton made a long series of experiment* of 

 a most interesting nature on the -iihtl" rehv 

 tjons between Lepidonterous lame and their 

 surroundings. The colours and markings have 

 a double source : ( 1 ) pigments derived from 

 the food-plants ; (2) pigments proper to the 

 larvae. A larva may be coloured from either or 

 both of these sources ; all greens seem due to the 

 chlorophyll, and most yellows to the xanthophyll 

 of plants; other colours to the proper pigment* 

 of the larvae. The derived pigments are more 

 frequently the basis of general resemblances to 

 surroundings, the true pigments of special and 

 detailed likeness. Poulton arranges the causes of 

 colour in larvae, in the presumed historic order of 

 their employment, thus : ( 1 ) Ready-made colour 

 in the internal tissues and organs, in the digestive 

 tract, fat, and dorsal blood-vessel; (2) derived 

 pigments which have passed through the walls of 

 the digestive tract into the blood or the tissue 

 under the cuticle ; ( 3 ) true pigment in the cuticle 

 and in the layer immediately oelow (the hypoder- 

 mis). But the point of most general interest is the 

 relation between the colour of the larvae and that 

 of their food-plants. Within the same species tin- 

 colours may vary to suit the colour of the feeding- 

 ground. Abundant instances of this are recorded 

 in Meldola's notes to Weismann's Studies. 

 McLachlan noted for instance that the larvae of 

 Eupithecia absynthiata were yellowish on the 

 yellow ragweed (Senecio jacobcea), reddish on the 

 purplish centaury (Centaurea niyra), and white 

 on the mayweed (Matricama). Poulton has the 

 credit of analysing this interesting relation. He 

 has shown that the influence of the food-plant 

 must act throughout a long period of larval life, 

 that the effects probably accumulate during suc- 

 cessive generations, and that the result cannot be 

 referred to the direct influence of the material 

 eaten. The interpretation is rendered particularly 

 difficult ' by the gradual working of the process, 

 often incomplete in a single life, by the excessively 

 complex and diverse result, and by the special 

 character of the stimulus, for it is only part of the 

 environment which produces any effect.' In the 

 case of the larva of Smerinthus ocellatiis, Poulton 

 has shown that the colour relation is adjustable 

 within the limits of a single life, and that the pre- 

 dominant colour of the plant is the inciting stimu- 

 lus. The colour adaptation is not in this case at 

 anyrate due to the gradual working of natural 

 selection, but to relatively immediate power 

 enabling the larva to suit itself to its conditions. 

 But the influence, though in one sense direct, is a 

 very subtle one. Poulton's investigations show 

 ( 1 ) that larvae have certain hereditarily trans- 

 mitted tendencies towards certain colours; (2) 

 that the colour of the leaf, and not the substance 

 eaten, is the agent which influences the larval 

 colours ; ( 3 ) that the influence is an intricate 

 nervous one, ' making itself felt by affecting the 

 absorption and production of pigments rather than 

 their modification when formed;' and (4) that 

 individual variations are comparatively unimport- 

 ant, though it is quite possible that variation 

 began somewhat uselessly in the pigments in the 

 blood, &c. , and were afterwards ' rendered effica- 

 cious by co-ordination with the environment.' 

 Some or Mr Poulton's most beautiful recent ex- 

 periments (1887) are those which show how the 

 golde^ surroundings of a gilt-lined box favour the 

 production of golden pupae. The above naturalist, 

 to whose observations this article is so much in- 

 debted, has done more than any one else to 

 penetrate into the physiological conditions of 

 caterpillar colour, but much still remains to b 



