Pratherodesmus, Shear, William A., Taylor, Steven J., Wynne, Judson & Krejca, Jean K., 2009

Shear, William A., Taylor, Steven J., Wynne, Judson & Krejca, Jean K., 2009, Cave millipeds of the United States. VIII. New genera and species of polydesmidan millipeds from caves in the southwestern United States (Diplopoda, Polydesmida, Macrosternodesmidae), Zootaxa 2151, pp. 47-65 : 48-49

Family Macrosternodesmidae Brölemann, 1916

Shear and Shelley (2007, 2008) diagnosed and discussed the Family Macrosternodesmidae , but the line of demarcation between the macrosternodesmids and nearctodesmids remains rather unclear, particularly in light of the fact that the genera described below have smooth metaterga, rather than having them divided into subquadrate elevated areas, as in Tidesmus Chamberlin, 1943 . Smooth metaterga, with few or no setae, are characteristic of nearctodesmids. The gonopods of at least two of the species of Pratherodesmus , new genus, have the process arising from the mediodistal side of the prefemur, called process B by Shear and Shelley (2007), reduced or absent, likewise the distal zone. The gonopods of macrosternodesmids and nearctodesmids are very similar, except that nearctodesmids have been maintained as having both process A and B, and macrosternodesmids only B, but in the present two genera, it would seem process A is indeed present, and as mentioned above, process B is absent in two species of Pratherodesmus . This blurring of the distinction between the families suggests to us that eventually they will need to be com bined under Macrosternodesmidae , the older name, as more taxa are discovered and described.

A recent paper by Djursvoll (2008) on the Iberian polydesmid genus Schizomeritus Verhoeff, 1931 , makes possible, through its clear SEM photographs and labeling, some reconciliation between the gonopod nomenclature originated by Shelley (1994) and used in previous papers on macrosternodesmids ( Shear & Shelley 2007, 2008), and that adapted by Djursvoll et al. (2001 [2000]) and Golovatch and Wytwer (2007). It appears that our process B is the exomere of Djursvoll (ex in our figures), process A corresponds to the endomerite (en in our figures), and our distal zone is Djursvoll’s tibiotarsus (tt in our figures). We previously have not used the term femorite, which Djursvoll (2008) uses for the part of the acropodite traversed by the seminal groove, but agree this term is a useful one, and apply it here. Our solenomere (s in our figures) and that of Djursvoll (2008) are the same. In the descriptions that follow we have decided to use the Djursvoll terminology, because it has the advantage of using terms that have already been in the literature for a considerable time, but now with clear definitions; application of this terminology allows for discussion of possible homologies between genera and families of the superfamilies Polydesmoidea and Trichopolydesmoidea. Earlier concerns expressed by Shear and Shelley (2007) now seem to us to be resolved.

Interestingly, the gonopod structure of species of Schizomeritus is very similar to that of the new genus Pratherodesmus , while the nonsexual characters are quite different. Species of Pratherodesmus have only small paranota, strongly produced posteriolaterally, and with regularly arranged rows of clavate setae on an otherwise featureless surface. Species of Schizomeritus are more typical of Polydesmidae , with large, squared, anteriorly elevated paranota not strongly produced posteriolaterally. The metazonite surfaces are deeply sculpted into bulging polygonal areas, and the setae are inconspicuous. On poriferous segments Pratherodesmus species have the ozopores nearly at the posterior corner and the edges of the paranota are not swollen, while in Schizomeritus the pores are more anterior and the edges strongly rebordered. Despite the similarity of the gonopods in general appearance, there are differences. The course of the seminal canal crosses over to the lateral side in Schizomeritus and ends in a vesicle and pulvillus. In Pratherodesmus , the canal stays mesal, does not end in a vesicle, and there is no pulvillus (although a dense group of cuticular fimbriae probably not homologous to a true pulvillus occurs). At this point it is not at all clear what the significance of these similarities and differences may be.

As previously remarked ( Shelley and Shear 2007), the millipeds of the superfamilies Trichopolydesmoidea and Polydesmoidea need close re-examination. Several families in both superfamilies, including Polydesmidae , remain poorly diagnosed and have been used as “wastebaskets” for enigmatic genera and species. We believe that at least two undiagnosed polydesmoid families are to be found in the very diverse, but largely undocumented, North American fauna, and have on hand many new generic and specieslevel taxa, primarily from the Pacific northwest. In the course of describing these taxa it may be possible to begin to make some sense of the two superfamilies, but without much more knowledge of small tropical polydesmidans, the composition and fate of the Trichopolydesmoidea remains uncertain.

Presently four genera from southwestern North America, including the two new ones below, have been assigned to the family Macrosternodesmidae . They may be separated by the following key: