Discussion

These specimens are members of the first broad group defined in the discussion of females of Nemesis lamna, i.e. small specimens (3.0 mm-5.0 mm) with a three-segmented abdomen and the fifth thoracic segment significantly smaller (25%-50%) than the fourth.

A number of described species come within this group.

In an attempt to separate these species I used the method described above to produce a ratio of widths, the cephalothorax being counted as 1, and the measurements placed in order of increasing relative width (see tables 3 and 4).

Table 4
Present material (females)

Host	Area	Seg. 2	Seg. 3	Seg. 4	Seg. 5	Gen. Seg.	Abd. 1	Total Length (mm)
Alopias	Probably	1.39	1.35	1.22	0.74	0.55	0.31	3.0
vulpinus	Cooks	1.47	1.56	1.39	0.69	0.54	0.29	4.6
	Strait	1.52	1.56	1.47	0.65	0.61	0.31	5.0
		1.54	1.59	1.37	0.71	0.58	0.27	3.6
		1.56	1.56	1.35	0.78	0.54	0.35	3.8
		1.56	1.61	1.39	0.69	0.59	0.31	3.85
		1.59	1.45	1.27	0.72	0.57	—	3.0
		1.64	1.69	1.54	0.72	0.59	0.30	5.0
		1.64	1.69	1.54	0.72	0.64	0.32	4.9
		1.64	1.69	1.54	0.86	0.59	0.33	4.3

The following characters distinguish the described females in this group of the genus Nemesis:

N. atlantica Wilson (1922, p. 60) has a comparatively broader cephalothorax than other described species and is also unique in having notches on the lateral margins of its second thoracic segment.

page 23

N. macrocephalus Shiino (1957, p. 392) and N. pilosus Pearse (1951, p. 362) are similar in having their fourth thoracic segment subequal in width to the cephalothorax (in other species, with the exception of N. atlantica, this segment is 33%, or more, wider than the cephalothorax) and also in having the spines on the second segment of the exopod of the second pereiopod subequal in length to this segment, in contrast to the much smaller spines described for other species. I am unable to separate these two species on the basis of descriptions in the literature.

N. versicolor Wilson (1913, p. 236) is unique in possessing a knob of each lateral margin of the cephalothorax.

N. pallida Wilson (1932, p. 464) and N. robusta (van Beneden) as described by Delamare Deboutteville and Nunes-Ruivo (1953, p. 211) differ mainly in the comparative width of their fifth thoracic segments. Considering the range of variation of this segment in the present material this character does not seem sufficient to regard these as separate species.

Cressey (1967, p. 6) described N. aggregatus which he separated from most other species on the basis of 10-12 small spines on the second segment of the second antenna, stating that these other species had 25-40 even smaller spines, and from N. tiburo and N. versicolor since in his species the row of very small spines, on the third segment of the second antenna of these species, was missing. In my material the second segment of this appendage has from a few to 39 very small spines, which vary considerably in size, although none are as large as those shown in his figure (p. 7, fig. 9). Also in my specimens the spination of the third segment varies from two to 21 very small spines, some of which may be arranged in one or two longitudinal rows. Under these circumstances the use of the spination of the second antenna as the sole criteria of taximonic distinctness must be considered doubtful until it has been investigated in further collection.

N. robusta (van Beneden) as described by Brian (1906, p. 72) is quite distinct from any described species of Nemesis in the narrowness of its cephalothorax which is about half the width of the second thoracic segment, while in other species it is two-thirds the width of this segment or wider. However, it seems likely, considering the usual accuracy of Brian's identifications and that similar specimens have not been recorded since, that the specimen figured was simply an abberrant individual.

N. tiburo Pearse (1952, p. 217) is described from a single, small (2.7 mm total length) specimen, without egg strings. It could easily be a juvenile form of one of the above species.

The present material clearly belongs to N. robusta as described by Wilson (1932) and Delamare Deboutteville and Nunes-Ruivo (1953).

Unfortunately van Beneden's original description and figures appear to refer to the male, and his figures (1851, pl. 3, figs. 1-2) show his specimen only in side view. However his specimens seem to correspond quite closely with the males in the present collection.

Previous records which appear to be of this species include:

Northeast atlantic—on Prionace glauca, Mustelus mustelus, Dasyatis pastinaca: Belgium (van Beneden, 1851, p. 97, 1870, pp. 4, 5 and 15).

on Alopias vulpinus: Brighton, England (Leigh-Sharpe, 1936, p. 410).

on Sphyrna sp.: Mauritania (Brian, 1924, p. 400).

on Cetorhinus maximus:^* Ayrshire, England (Jenkins, 1936, p. 316).

page 24

Southeast atlantic—on Sphyrna zygaena and Mustelus sp.: Angola (Nunes-Ruivo, 1956 ^*).

on Alopias vulpinus: Table Bay, South Africa (Barnard, 1948, p. 250).

Mediterranean—on Prionace glauca: Sète (Delamare Deboutteville and Nunes-Ruivo, 1953, p. 211).

on Mustelus mustelus, Dasyatis aspersa, Raja oxyrincha, and R. macrorhynchus: Adriatic (Valle, 1880, p. 67).

on Hexanchus griseus: Adriatic (Valle, 1884, p. 1).

Northwest atlantic—on Alopias vulpinus, Carcharhinus milberti, C. obscurus, Carcharodon carcharias, Odontaspis taurus, Galeocerdo cuvieri: Marthas Vineyard, Mass. (Wilson, 1932, p. 464).

on Carcharhinus limbatus: Beaufort, Carolina (Pearse, 1947, p. 9).

on C. obscurus, Rizoprionodon terraenovae, Sphyrna sp., S. tiburo, smooth-toothed shark (Aprionodon isodon (?)), spot tipped ground shark: Lemon Bay, Florida (Bere, 1936, p. 604).

on C. limbatus: Port Aransas, Texas (Causey, 1953, p. 14).

on Negaprion brevirostris: Dry Tortugas (Wilson, 1935, p. 340).

So far as I am aware, this is the first record of this species outside the Atlantic and Mediterranean.