44 



MEMOIRS OF THE NATIONAL ACADEMY OF SCIENCES. 



comparatively small size of the ch'peus, which is not sunken below the level of the epicranium 

 as it is in the S_ymbonibycina (compare XLIV tigs. 10, 11, 12, with tigs. 13 and 14), the clypeus 

 being still smaller in Adelocephala than in the Heterocampinie. 



As regards the armature, that of the notodontian groups Notodontina- and Heterocampinaj 

 is closely' similar to that of Ceratocampime. In neither group does more than a single seta arise 

 from a tubercle. 



In Schizum concinna (fig. 3) the tubercles are solid, chitinous, forming horns; the position 

 of the tubercles is in general much as in Adelocephala, l)ut, as we should expect, the armature 

 is more primitive; tubercle Hi. is near the spiracle, io moved up to a position just below and 

 behind the spiracle; while /■ is quite remote, and ri nearly midway between t' and vil. Their 

 position, especially that of lit., U\ and i\ is the same in Iletervcampa giiftivitta, stage I, Seirodonta 

 hllhuata, and presumable in Notodontidffi in general. On the other hand in the Ceratocampidaj 

 tubercles iv and v are united. The difference is a family one, but this does not militate against 



Fig. 3. — Head nnd first five trunk segments of ScJiizura concmna, showing the 

 iirrangement of tubercles i-rii. 



the derivation by rapid evolution (tachygenesis) of the more specialized or modified Cerato- 

 campinifi from the Notodontid;*?, as it is now a matter of little doubt that evolution fi-oni one 

 family or order or class to another may have in most cases at least been effected by a jump or 

 sudden nuitation, without a long series of connecting links. 



It should also be observed that in the Heterocampina? we have the frequent concurrence of 

 '"horns," i. e., of the conversion of tuliercles into solid chitinized horn-like processes, bearing a 

 seta at the end. While this may l)e a sporadic specialization of the dorsal tubercles of stage I of 

 from only one pair (prothoracic in //. hiundata and II. unicolor) to as many as seven [IF. (iJiliqiut) 

 and nine {11. guttiivftK), yet it is not without significance, as ]:)ointingto the evolution of a group 

 like the Ceratocampida', where they are retained throughout larval life. 



Here the question arises whether these antlers and spines of Heterocampa and the reduced 

 prothoracic horns of Mncrnnx-aDipd inarthcxia and of Cerura may not have been handed down 

 from the genus Schizura, or at least that section of it, or the incipient genus represented \)X 

 S. cnncinna. in which ail the segments bear tul)ercles (/), which have become specialized into 

 stout spines. 



While the pupa of Heterocampa resembles that of Eacles in the general sha])e of the head- 

 region, and in having a forked cremaster, what appears to be a difference of family importance 

 is seen in the two prominent divisions of the lower part of the front of the head, representing 

 the clypeus and lul)rum. 



