716 PBOCEEDJXGS OF THE NATIONAL MUSEUM. vol. xxix. 



other; and, on the other hand, anions- the tvpieal l^deozoic bhittoid 

 forms there are also those in which tlie type of venation prevailing- 

 today is to be observed. In primordial tinie, the hind wings of blat- 

 toids were alread\" straight as at present, exhibiting an anal area 

 plaited lengthwise ((-(Mitrarv to the view of Sellards): there were also 

 even then forms in which a cross folding of the wing was indicated (in 

 the European Permian), and; as a rule, cross veins were clearly 

 developed. In many living forms the cerci are still long and dis- 

 tinctly jointed. The ovipositors mentioned Ijy Sellards could probal)ly 

 not hold ground in a critical investigation, and may in all probability 

 never have existed; they have been hitherto observed only in several 

 nymphs, which very likely l)elong to the protobhittoids, but as yet 

 in no true blattoid imago, and it seems to me very hazardous to assume 

 the existence of long ovipositors as a character of the raheoblattaria_\" 

 On the other hand, the discovery of several e^g cases proves to us 

 that the Carboniferous blattids even at that time laid their eggs in a 

 way similar to that which their descendants still practice to-day. The 

 young stages of Paleozoic blattoids also strikingly resemble those of 

 recent forms, though in general it is to be noted that in individual 

 cases the former, in their more distinctly jointed and longer cerci and 

 in their more slender form, more ninirly approach the type of palaio- 

 dictyopteran lai'va'. As pre\'iously mentioned, it is extremely diffi- 

 cult to make a sharp distinction between the protobhittoids and the 

 blattoids, and at the present time one can hold only to the fact that 

 the former, at least in respect to the venation of their wings, are nuich 

 more closely allied to the primitixe type (Paheodictyoptera) than the 

 latter. 



A systematic arrangement of Paleozoic blattoids in natural groups 

 clearly meets with not inconsiderable difficulties, because in the course 

 of time all series must be bound together by intermediate forms. The 

 systematic arrangement attempted by Scudder has proved itself 

 whoU}^ defective in every respect, and rests upon entirely artiticial, 

 arbitrarily selected characters. Moreover, as a rule, Scudder's generic 

 diagnoses do not at all apply to the majority of forms as arranged by 

 him, and according to this system very closely related sijecies must ))q 

 separated in widely different genera. 



I have therefore attempted to set up a new grouping, to the extent 

 of bringing the genera and families, as far as possible, into agreement 

 with those of recent blattoids. In so doing, I haAC been forced to 

 erect a large number of new groups, in order to avoid uniting hetero- 

 geneous elements. I am fidly convinced that many of my genera will 

 be combined when more abundant material becomes known; still I 



« The ovipositors mentioned by Brongiiiart as occurring in several Carboniferous 

 blattids are likewise a hisim indnnr, and no "prolongation of the lower genital 

 process." 



