19] LEPIDOPTEROUS LARVAE — FRACKER 19 



A third position might be taken with regard to development by- 

 successive molts on the supposition that it might not constitute an 

 ontogeny in the usual meaning of the term. At the same time it must 

 be recognized that the instars through which a caterpillar passes are 

 as necessary a part of its development as are the changes within the egg. 

 Specific evidence of recapitulation in the life history of animals which 

 molt is taken up in a later paragraph. 



Constructive evidence on the recapitulation theory as applied to 

 larval instars is considerable in amount. Some of it has been sug- 

 gested in answering the objections and merely the outline of this evidence 

 is given below. There is no need to develop the different points. 



I. The instars of other Arthropoda recapitulate their ancestral 

 history. Examples: Sacculina and its degeneration; the changes of 

 barnacles ; the zooea, mysis stages, etc., of Decapoda. 



II. The phylogeny of other insects is shown by their postembryonie 

 development. Examples: in Coccidae, the presence of the anal ring in 

 the nymphs of Kermesinae, and the appearance of primary and secon- 

 dary pygidia in certain Diaspinae ; in Coleoptera, the campodeiform 

 larva preceding the cruciform in certain cases. 



III. The development of the setal plan of lepidopterous larvae 

 itself bears prima facie evidence that it constitutes a recapitulation. 

 1. All the Frenatae are almost identical in the first instar when that 

 stage is different from later ones. 2, The first stage of the Jugatae is 

 much nearer that of the Frenatae than later stages are. 3. Larvae with 

 tufted setae, as arctians, usually possess only the primary setae before 

 the first molt and these are in the typical position. 4. The armature 

 of the larvae of Nymphalidae and certain other butterflies or specialized 

 Lepidoptera is not homologous with that of the moths but is preceded 

 in the first stage by the setae in the typical position. 5. The absence 

 of theta on all the segments of Hepialus and the Frenatae in the first 

 stage, and its presence on all the segments of Hepialus and on the thorax 

 of Frenatae in the next stage, can have no other meaning than that this 

 seta is a later arrival than those which are present from the time of 

 hatching and that it has become established in the Frenatae on the thorax 

 only. 6. Setae are distinct in the first instar of Sphingidae, Dioptidae^ 

 and other groups, but are very much reduced or wanting later. At the 

 same time their descent from forms in which the setae are distinct is 

 unquestionable. 



The writer has become convinced from many facts of which the 

 preceding are only examples, that the first-stage larva of Lepidoptera 

 represents the ancestral type; that the arrangement of the setae in this 

 instar is essentially the same their ancestors bore in some remote 



