120 Sellards — Structure of Paleozoic Cockroaches, 



Abdomen. — Ten terga are to be seen on the abdomen of both 

 male and female. The projecting edges of the terga are well 

 developed from the third or fourth to the ninth. The tenth 

 tergum is smaller and more or less rounded. The two cerci 

 project from beneath it. The seventh, eighth, and ninth 

 terga are well developed, presenting in this respect a marked 

 contrast to the common existing cockroaches, among which the 

 corresponding terga in both male and female are very much 

 reduced and partly covered by the seventh, even in the later 

 nymph stages. More difficulty is encountered in the study of 

 the sterna. They are seen either in outline through the terga, 

 or, the terga having been removed, are viewed directly. The 

 first sternum, which was doubtless, as in living forms, much 

 reduced and imperfectly chitinized, has not been observed. 

 The third to the ninth can be seen lying beneath their respec- 

 tive terga on a few individuals, presumably males. The 

 seventh sternum of the female is enlarged, rounded, and lies 

 beneath the seventh, eighth, and a part of the ninth terga. 

 The relative position of the parts can be determined from 

 several specimens (Figure 12). The position of the eighth 

 and ninth sterna of the female has not been determined. As 

 Brongniart has suggested (Hist. Ins. Foss., p. 417), the habit 

 of depositing the eggs within an egg case was probably not 

 common at that time. Many of the species doubtless deposited 

 their eggs singly either on the ground or underneath the bark of 

 trees or within small stems. Certain paleozoic ferns in imme- 

 diate association with cockroaches have been observed to pre- 

 sent a row of slits along the rachis, which appear to have been 

 made by such an organ as the ovipositor of the Paleoblattidse, 

 and, as has been stated in an earlier paper, are indeed strik- 

 ingly similar to scars seen on the stem of the common Amor- 

 jpha fruticosa (false indigo) and said to be made by katydids.* 

 Nevertheless some genera may have acquired the habit of put- 

 ting the eggs within egg cases as early as the latter part of the 

 Carboniferous. The writer has recently obtained a fossil from 

 the Upper Coal Measures of Kansas which has a striking 

 resemblance to the egg cases of modern cockroaches (Figure 25). 



Ovipositor. — The ovipositor is present on several specimens. 

 The parts of this organ in one very young individual of 

 Etoblattina, mazona (Figure 11) have apparently not yet 

 united, and present a striking similarity to the early stages 

 in the development of the ovipositor in the Locustidae, as 



* A description of the scars found on the rachis of Tceniojoteris Brongniart 

 and Glenopteris Sellards has been given by the writer in the Kansas Univ. 

 Quart., vol, ix, p. 184, July, 1900 ; ibid., vol. x, pp. 9-12, Jan., 1901. Simi- 

 lar scars on Tceniopteris from West Virginia are described by Fontaine and 

 I. C. White (Perm. Flora, p. 92) ; and on Macrotceniopteris from Virginia by 

 Fontaine (Mon. VI, IT. S. G. S., Older Mes. Flora, p. 18, 1883). 



