NATURE 



457 



THURSDAY, SEPTEMBER 15, 1881 



THE THEORY OF DESCENT 

 Studies in the T/teo?y of Descen/. By Dr. Aug. Weismann, 

 Professor in the University of Freiburg. Translated 

 and Edited by Raphael Meldola, F.C.S. Part II. On 

 the Origin of the Markings of Caterpillars. Cn Phyletic 

 Parallelism in Metamorphic Species. With Six Coloured 

 Plates. (London ; Sampson Low, Marston, Searle, and 

 Rivington, 1881.) 



THE first part of this work, devoted to an examination 

 of the phenomena of seasonal dimorphism in butter- 

 flies, was noticed a little more than a year ago (Nature, 

 vol. xxii. p. 141). We now have a second instalment of 

 much greater bulk, comprising two separate essays. In 

 the first and most important of these Dr. Weismann gives 

 us the results of a detailed study of the changes in the 

 markings of the caterpillars of the Sphingidffi during the 

 course of their growth and development, and enters at 

 great length into the various questions to which the 

 phenomena observed give rise. Accepting the doctrine 

 that the ontogeny or development of the individual gives 

 us a more or less accurate notion of the phytogeny or 

 course of development of the race, he endeavours with 

 some success to determine the ancestral forms of the 

 various genera of the Sphingids by means of the succes- 

 sive changes of form and coloration of the larvre. The 

 main facts which he has here established are, that all the 

 larvre are born of a uniform tint — that the first markings 

 are longitudinal lines — that the oblique lines when they 

 exist always appear later, and the ringed or ocellated 

 spots last of all. Great changes of colour also occur in 

 some species, but aU the more important changes, whether 

 of colour or marking, only take place after the larvae have 

 acquired a considerable size. From the whole assemblage 

 of facts in this branch of the inquiry he deduces the 

 following three laws of development : — 



" I. The development commences with a state of sim- 

 plicity, and advances gradually to one of complexity. 



'' 2. New characters first make their appearance in the 

 last stage of the ontogeny. 



"3. Such characters then become gradually carried 

 back to the earlier ontogenetic stages, thus displacing the 

 older characters until the latter disappear completely." 



These laws are liable to be modified in various ways by 

 the influence of natural selection, and especially by the 

 need for protection, whence arise the various markings of 

 the different groups, and the peculiar divergences often 

 noted in their development at corresponding ages. 



Having thus established the general developmental 

 history of the markings of caterpillars, and explained by 

 a few simple principles the chief anomalies they present, 

 Dr. Weismann passes on to the still more interesting 

 inquiry as to the biological value or actual meaning and 

 use of the markings in each individual case. He first 

 shows that colour itself, irrespective of marking, has a 

 distinct biological value, being always either protective or 

 a signal of uneatableness. The Sphinx larvse when young 

 are almost always green, resembling the leaves of the 

 plants on which they feed and rest. When they get 

 larger, however, they frequently change to brown, and 

 this change is always accompanied by a change in habits. 

 Vol. XXIV. — No. 621 



the insect feeding at night, but during the day concealing 

 itself on the ground or amidst dead leaves and branches. 

 This occurs chiefly among the species -nhich feed on 

 low shrubs or herbs and can thus easily descend to 

 the ground to conceal themselves during the day ; while 

 those which feed on large trees acquire markings which 

 assimilate them more completely to the foliage cr flowers 

 which surround them. The simplest form of marking — 

 longitudinal stripes — is common on all caterpillars which 

 feed on grasses or other plants in which straight lines are 

 a prevalent feature, and this style of marking is that 

 which usually appears in the young sphinx larvae. But 

 as they grow larger diagonal stripes or bands variously 

 tinted or shaded appear, and this style of marking is 

 found to assimilate so well with the oblique veining of the 

 leaves that the caterpillars are very difficult to sec when 

 resting among them. This is the case even when the 

 oblique lines are margined with violet or other bright 

 colours, since, however conspicuous these markings may 

 be when the insect is examined in captivity, they are 

 found to blend perfectly with the lights and shadows of 

 the foliage which surround it in its natural habitat. As 

 an example we have the following account of the brilliantly 

 coloured larva of the Death's-Head Moth on one of its 

 natural food-plants : — 



"At Cadiz on the hot, sandy shore, Solannm violaceum 

 grows to the height of three feet, and on a single plant I 

 often found more than a dozen Atropos lar\-3e resting with 

 the head retracted. It can easily be understood why the 

 lateral stripes are blue when one has seen the South 

 European Sotancre, on which this larva is at home. 

 Sotantimviolaceum is scarcely green : violet tints alternate 

 with brown, green, and yellow over the whole plant, and 

 between these appear the yellow-anthered flowers, and 

 golden yellow berries the size of a greengage. Thus t 

 happens that the numerous thorns, an inch long, between 

 which the raterpillar rests on the stem, pass from violet 

 into shades of blue, red, green, and yellow." 



Many of the adult sphinx larva; however are adorned 

 with ring-spots or eye-spots, and these have been founrl 

 to serve two distinct purposes. Sometimes they occur on 

 several of the segments, but of slightly different sizes, as 

 in a North American species {Cho'roeampa tersd), the red 

 spots on which imitate the small red flowers of the plant 

 on which it feeds ; while in the European Deilephila 

 hippophaes, the grey-green larva with orange-red spots so 

 exactly assimilates to the foliage and fruit of the sea- 

 buckthorn on which it feeds, that Dr. Weismann has 

 often shown to people as many as six or eight of the large 

 caterpillars on one buckthorn branch without their being 

 able at once to detect them. In other cases we find very 

 large eye-spots on the fourth or fifth segments only, 

 coupled with the habit of retracting the head and first 

 three segments, so that an appearance is produced of a 

 broad head with two very conspicuous eyes. Whenever 

 the insect is disturbed it thus retracts itself, and it has 

 often been conjectured that the effect is to frighten away 

 its enemies. This Dr. Weismann has proved to be actu- 

 ally the case. On placing the larva of the Elephant 

 Hawk-Moth {Chcerocmnpa Elpe}ior) in a trough used for 

 feeding fowls, a number of sparrows and chaffinches flew 

 down from the neighbouring trees to pick up some stray 

 food. 



" One bird soon flew on to the edge of the trough, and 



