Trichozonus Carl, 1905

Golovatch, S. I., Fiemapong, A. R. Nzoko, Tamesse, J. L., Mauries, J. - P. & VandenSpiegel, D., 2018, Trichopolydesmidae from Cameroon, 1: The genus Hemisphaeroparia Schubart, 1955. With a genus-level reclassification of Afrotropical genera of the family (Diplopoda, Polydesmida), ZooKeys 785, pp. 49-98 : 54-56

Trichozonus Carl, 1905

Type-species.

Trichozonus escalerae Carl, 1905, by monotypy; Equatorial Guinea.

Female.

20 segments, body length 8 mm; paraterga modest, tergal setae long and bacilliform. Like Bactrodesmus , this genus remains too poorly documented to consider in our following analysis.

The above brief accounts are given to reiterate the foundations of the previous classification and to offer a new one below. The classification developed by Mauriès and Heymer (1996) in their review of the African Trichopolydesmidae (= Fuhrmannodesmidae ) cannot be accepted because it totally lacks any evolutionary perspective. These authors themselves admit their approach as being “bastard” and overly lumping, as they placed all species but one ( Bactrodesmus , dubious) in the single genus Sphaeroparia , within which they accepted six subgenera and suggested several synonymies.

Golovatch (1992, 1994) provided an evolutionary scenario for the genera of Trichopolydesmidae (= Fuhrmannodesmidae ) known from South America, accepting as the basalmost those genera showing rather small, narrowly fused1, subglobose gonopodal coxae that form no significant gonocoel in which to hinge the largely exposed, usually rather complex and elongate telopodites. Amongst the Afrotropical trichopolydesmids such are the genera Elgonicola , Heterosphaeroparia , Megaloparia , and Sphaeroparia (Figs 1D, 2A, 1B, and 1A, respectively). At the opposite end which obviously represents the evolutionary summit are such genera as Dendrobrachypus , Eburodesmus , and Mecistoparia (Figs 2D, E and 2B-C, respectively). Their gonopodal coxae are especially voluminous and inflated laterad; the particularly deep gonocoel is capable of concealing entire or nearly entire telopodites. Sometimes the coxa shows a rather conspicuous apicolateral process or lobe, this being suberect ( Heterosphaeroparia , Figure 2A, cp) or directed more or less mesally ( Eburodesmus , Figure 2E, cp). Schubart (1955) created his Eburodesmus , based on such evident and more or less mesally directed coxal processes (Figure 2E, cp), but we think this may be only a species-level character occasionally present also in distinctly more basal taxa, e.g., Heterosphaeroparia (Figure 2A, cp). A series of transitions can be seen between the two extremes, e.g., in Heterosphaeroparia , Elgonicola or Hemisphaeroparia , when 1-3 prominent branches project well beyond the coxal margin, while the telopodite is mainly deeply sunken inside the coxa. Typically there are 2+2 particularly strong setae near the place of central fusion of both coxae, while their lateral surfaces are normally more or less clearly granulate/alveolate and setose.

The cannula in Afrotropical Trichopolydesmidae is invariably medium-sized, tube-shaped, strongly curved, long, and slender, its tip entering the densely setose, funnel-shaped, “prefemoral” part of the telopodite which extends from base to apex. The seminal groove is mostly rather short and straight, usually running on the mesal side to end on a simple, more or less finger- or spine-shaped, sometimes retrorse solenomere (sl). In a few cases, the solenomere ends near a kind of hairy pulvillus or shows a tooth at its base. However, there is a remarkable exception, when the seminal groove makes a distinct loop before proceeding onto a prominent solenomere (e.g., Eburodesmus , Figure 2E). Such situations seem to imply telopodite torsion. The length of a solenomere is usually barely more than species-specific, but the course of the seminal groove is definitely a generic-level character, as can be seen in Neotropical or Southeast Asian Trichopolydesmidae ( Golovatch 1992, 1994, Golovatch et al. 2014, Golovatch and VandenSpiegel 2016).

The gonopodal telopodite is typically helmet- or boat-shaped. At the apicolateral end which is opposite to the “prefemoral” funnel there is a group of partly especially stiff setae usually extending basad across or over most of the funnel. Variations in the shape of the remaining parts of the telopodite are especially prominent. Among the most common outgrowths or processes of the telopodite we choose to denominate the following. A unipartite telopodite which shows only a single prominent branch, apical in position (ab), is observed in Dendrobrachypus , Mabocus , Mecistoparia and Physetoparia (Figs 2D, 1E-F, 2 B–C and 1C, respectively). Bi- or tripartite telopodites are more usual, sometimes also supplied with evident central lobes (lo). The branch which is the closest to the apex is denominated an apical branch (ab), while that lying the closest to the basal part is termed a basal branch (bb). If there is another branch located between ab and bb, this is a medial branch (mb). Two of or all three branches are often strongly adjacent to each other. A few species groups may be provisionally outlined based on the general conformation of the gonopodal telopodites.

Two different general trends can be observed in the evolution of the gonopods in Afrotropical Trichopolydesmidae . While the coxa tends to grow and develops an increasingly prominent gonocoel, the originally relatively complex and largely exposed telopodite, at least in some cases, seems to shrink gradually and often becomes simplified. Indeed, the presumably basal taxa with particularly small gonocoxae (e.g., Sphaeroparia , Elgonicola or Megaloparia , Figs 1A, 1D and 1B, respectively) show complex apical structures on telopodites. In cases like Eburodesmus (Figure 2E) and Heterosphaeroparia (Figure 2), it is the coxal apicolateral process (cp) that takes the protective function to secure the still complex and exposed parts of the telopodite. Usually, when the outgrowths are mostly concealed and little exposed beyond the coxa, they are particularly few and sometimes reduced just to one branch (e.g., Dendrobrachypus , Figure 2 D or Mecistoparia , Figure 2B, C). Interestingly, the single species of Sphaeroparia , S. simplex Golovatch, 2013, described from the Balkans and several Greek islands ( Golovatch 2013), i.e., well beyond the Afrotropical realm, shows voluminous gonopodal coxae and especially simple and small telopodites, these latter being mostly concealed inside the gonocoel. Now that Sphaeroparia receives a clear definition (see below) and joins the basal group of genera with still small gonocoxae and complex and well-exposed telopodites, the identity and generic allocation of S. simplex must be revised (Vagalinski et al., in preparation).

When a delicate solenomere is left well-exposed (Figs 1C, 2D, F-G), it appears to always be protected by adjacent stronger structures serving as a solenophore. The number and shapes of such accessory outgrowths (processes and lobes) varies between species, but they are always few (1-3) and tend to be little exposed to fully concealed or even absent in more advanced genera.

Such are the main guidelines, all based solely on gonopodal anatomy, to follow in order to obtain new generic delimitations arranged according to an increasing complexity of the coxae, combined with a decreasing complexity of the telopodites. Somatic characters such as the number of body segments (19 in the male or in both sexes, or 20 in both sexes), the pore formula (always normal, but can be a little abbreviated on the last 1-2 segments before the telson: 5, 7, 9, 10, 12, 13, 15-17(18,19)), the degree of development of paraterga (usually moderate, often small, but never really well-developed and strongly flattened dorsally), the position of the ozopores (usually open flush dorsally near the caudal corner of pore-bearing paraterga), the shape of tergal setae (short and clavate to very long and bacilliform), the presence of modifications on male legs 2 (enlarged in Bactrodesmus ) and on the male head (ranging sporadically from nothing to a strong, central bulge or a mushroom-like or bulbous tubercle), they all are considered here as species-specific. This situation agrees with general wisdom derived from other tropical faunas ( Golovatch 1994, Golovatch et al. 2014).

A somatic character that deserves special mention is the unusually strongly developed peri-spiracular structure on segment 2. In all new species described below, the spiracle is located on a high finger with a complex tip placed next to coxa 2 (Figs 11K, 21J-L). Is this a generic feature that characterizes Hemisphaeroparia ? Could Cook (1896b) have mistaken the spiracle for leg 2 in his Bactrodesmus ?

The generic reclassification presented below considers only the type species, leaving aside the other component species and their allocations to our next paper on Afrotropical Trichopolydesmidae (Golovatch et al., in preparation). Two nominal genera, Bactrodesmus and Trichozonus , are dubious and are for the time being to be discarded from further consideration, because their gonopodal characters remain unknown. Whereas B. claviger is generally possible to revise or recognize based either on type or topotypic material from Liberia, since male legs 2 of this species are said to be conspicuously enlarged ( Cook 1896b), the identity of T. escalerae is bound to remain obscure ( Carl 1905) until a male topotypic sample from Fernando Po becomes available for study.