406 ENTOMOLOGICAL NEWS. [Nov.,'17 



the two groups in this respect. On the other hand the general ap- 

 pearance of the region abc of the Termite (Fig. 3) is much 

 more hke that of a Phasmid (Fig. 8) or Plecopteron (Fig. 7). 

 A study of the head, neck and prothoracic regions would thus 

 indicate that the Termites are quite closely related to the Blat- 

 tids, but have retained many characters suggestive of affinities 

 with the other groups studied, and the Termites may thus be 

 regarded as occupying a position somewhat intermediate be- 

 tween the Blattids and the other groups. 



It would be inadvisable to base one's conclusions as to the 

 relationships of the lower insects on a study of the head, neck 

 and prothoracic regions alone, and the present paper is there- 

 fore but one of a series in which the different body regions 

 have been compared part by part in the different groups ; but 

 1 am hoping to show that a study of the other structures will 

 in a large measure bear out the conclusions to be drawn from 

 the regions here discussed. Other investigators would derive 

 the Orthopteroid insects from Blattid-like forebears, and would 

 also trace the Grylloblattid line of development back to a Blat- 

 toid ancestry ; but it is only fair to demand that they shall 

 produce equally convincing proof of their contentions, which 

 should be drawn from the facts of comparative anatomy, since 

 comparative anatomy, after all, furnishes us with the most 

 reliable evidence of relationships, and is, in fact, the main- 

 spring of all systematic work ! 



Before leaving the subject of the head structures, I would call 

 attention to the antennif er, or antenna-bearing process Af which 

 is usually situated midway up the outer portion of the antennal 

 ring An in the Grylloblattids, Forficulids, etc. (Figs. 4, 5, 

 etc.), while in the Blattids (Fig. 2) it has migrated mesalward, 

 and in most of the members of this group (e. g. Ectobia, etc.) 

 this antenna-bearing process is usually located higher up along 

 the median portion of the antennal ring (i. e. on the side of the 

 antenna toward the median line of the head). The position of 

 the antennifer, however, is not sufficiently constant to be of 

 any great value in determining the relationships of the groups 

 in question. 



