PROM EGG TO INSECT SNODGRASS 399 



From all the evidence bearing on the segments and the appendages, 

 we may conclude, then, that the ancestral insects were creatures 

 somewhat like centipedes, consisting of a series of segments having 

 each a pair of appendages, and that modern insects have evolved 

 from this simple form to their present complex structure by a defi- 

 nite grouping of their segments and a specialization of certain ap- 

 pendages, accompanied by a loss of those appendages that were 

 found to be useless. This is not to say that insects are descended 

 from centipedes. The centipedes have had their own line of descent 

 from ancestors, which, perhaps, were the same as those of the insects. 

 The insects have departed farther from the primitive type than have 

 the centipedes. 



THE EMBRYONIC COVERINGS 



The embryos of most insects become protected in one way or an- 

 other from the egg shell during a part or all of their period of de- 

 velopment. With some of the more primitive species the germ band 

 doubles upon itself ventrally at an early stage and sinks into the yolk 

 in the form of an inverted U, in which position it may remain until 

 the embryo is almost fully formed. A simple procedure of this sort 

 occurs with the little insects known as springtails, an embryo of 

 which is shown at A of Figure 16, taken from the work of Philip- 

 tschenko. It is significant to note that this method is characteristic 

 also of centipede embryos and of the embryos of many arthropods 

 other than insects. Another of the lower insects, the fish moth, or 

 more entomologically, Lepisma, begins its development on the sur- 

 face of the egg in the usual manner, but it too soon curves and sinks 

 into the yolk (fig. 16 B), as shown in an account of its development 

 by Heymons. Lepisma, however, sinks so deeply into the egg that a 

 part of the surrounding thin blastoderm is turned in also, and the 

 mouth of the cavity closes to a small pore. Later, the Lepisma em- 

 bryo partially emerges and completes its development with its head 

 and thoracic parts again on the surface (C). Many insects that de- 

 velop entirely on the surface have the tail end bent forward as if they 

 had, at some time in their past history, likewise partiall}^ emerged 

 from a sunken position. 



Again, there are embryos of other orders of insects that turn into 

 the yolk tail end first (D) and stretch out within it in a reversed and 

 inverted position (E), in which position they remain until some 

 time before they are ready to hatch, when they revolve back to the 

 surface of the egg (F). 



The usual style of embryonic protection among insects, however, 

 is by means of covering membranes that grow over the embryo from 

 the surrounding thin part of the blastoderm. In typical cases of this 

 kind a fold of the blastoderm erows out around the sides of the jrerm 



