THE PROBLEMS OF TRANSFORMATION 261 



The common origin of these orders is indicated also by the 

 close resemblance between the pupae of various groups. The 

 " free " type of pupa is characteristic of all Coleoptera, all 

 Hymenoptera, all Mecoptera, Neuroptera and Trichoptera, 

 and of the most primitive among the Lepidoptera and 

 Diptera. Within these two last-named orders we notice such 

 specializations as the fully obtect pupa of the higher moths 

 and butterflies, and the puparium of the muscoid flies. 

 These specializations within the limits of an order make the 

 general similarity of pupal form throughout the Endopterygota 

 all the more convincing as evidence of true relationship. 



THE HISTORY OF THE INSECT ORDERS 



The comparison of larval forms characteristic of various 

 insect groups, has led us to conclude that they may be 

 derived from a primitive type of armoured larva with 

 limbs on most of the abdominal segments in addition to the 

 thoracic legs ; such a larva can be compared with some of the 

 primitive wingless insects (Apterygota). We have seen that 

 these Apterygota, as well as the primitive larvae of the mayflies, 

 show in their structure affinity to the Crustacea, and with 

 differences of opinion on points of detail, there has been during 

 recent years an increasing tendency 1 to regard Insects and 

 Crustacea as truly related. The growth of insect-wings has 

 been a recurring theme in this book ; our knowledge as to the 

 origin of wings in the whole insectan class is scanty and un- 

 certain, because insects as a class are not abundant as fossils, 2 

 and the earliest insects, whose remains are preserved to us 

 entombed in European and American rocks of Upper Carboni- 



1 E. R. Lankester : " The Structure and Classification of the Arthro- 

 poda ". Quart. Journ. Microsc. Sci., XLVII. 1904. G. H. Carpenter : 

 " On the Relationship between the Classes of the Arthropoda ". Proc. R. 

 Irish A cad., XXIV. 1903. " Notes on the Segmentation and Phylogeny 

 of the Arthropoda ". Quart Journ. Micr. Sci., XLIX. 1905. 



2 For readers unfamiliar with geological nomenclature, the following list 

 of names given to the various divisions of stratified rocks, as distinguished by 

 means of the fossils that they contain, may be useful. The systems are 



