Ischnoptera rufa occidentalis Saussure 1862

Ischnoptera rufa occidentalis Saussure (Plate IV, figures 9 and 10.)
1862. I[schnoptcra] occidciilalis Saussure, Rev. et Mag. de Zool., (2), xiv, p. 170.
[ 9 ; New [Orleans, Louisiana].] 1893. Ischnoptera coriformis Saussure and Zehntner, Biol. Cent. -Am., Orlh., i,
p. 37, pi. iii, fig. 25. [ 9 , Nicaragua.]
Tabernilla, Canal Zone, Panama, (Busck), i cf .
Cabima, Pan., V, 19, 191 1, (Busck), i 9.
Chorrera, Pan., V, 14, 1912, (Busck), i 9.
These specimens show not only the highest development in the present race as to size, but also in intensification of coloration and in the diagnostic features of the male genitalia. It is therefore evident that we ha'e in Panama no intergradation between rufa rufa and rufa occidentalis, but instead the most divergent type developed in the latter plastic race, contrasting strongly with the constant rufa rufa.
The distribution of rufa occidentalis is now known to extend in continental North America from the subtropical regions of the Gulf of Mexico southward to Panama; that of rufa rufa from the northeastern portion of South America westward to Panama and northward throughout the West Indies. It is possible that rufa originally passed from the West Indies to continental North America, and that the area where intergradation between rufa rufa and rufa occidentalis would occur no longer exists. If so, we would find in Panama the western limit of the distribution of typical rufa and the most southern point reached by rufa occidentalis. In this way alone do we feel able to account for the signal differentiation between these two races in the same region. The convergence in the large series of rufa occidentalis before us toward typical rufa shows definitely that these two types represent races, and nothing more, of one and the same species.
The material here studied of rufa occidentalis is in every way similar to specimens from Pozo Aziil, a locality in the southwestern
lowlands of Costa Rica, about ten miles up the Rio Pirris.^"^ This optimum type probably represents an incipient geographic race, but as yet has not reached a sufficient degree of differentiation to warrant nominal recognition.
That the present insect, though not domiciliary, thrives about human habitations under litter of all kinds, must be considered in studying material. It is this factor which increases the possibility of specimens being found adventive and not peculiar to the locality at which they were secured.
The concealed male genitalia in this species are very complex. The genital hook is simple, the titillator specialized distad (plate IV, fig. 9), this specialization more decided in riifa debilis Hebard and ruja occidentalis than in ruja rufa. The paired plate beneath the supra-anal plate is highly specialized on both sides, developed into chitinous lobes and shafts, of which the sinistro-dorsal shows the greatest specialization. The usual development of this shaft is here shown (plate IV, fig. 10). Individual variation occurs in which one or both of the marginal teeth disappear, and in ruja occidentalis the shaft occasionally is much broader, or much more elongate and slender. In rufa occidentalis the ventral surface of the supra-anal plate frequently bears a large, irregularly rounded, mesal lobe covered with stiff bristles, or is unspecialized as is normal in ruja rufa. These features of variability in the male concealed genitalia are a very unusual occurrence in the Blattidae.