Page 14 - Korn_alii_2006

Sister species within Triops cancriformis • M. Korn et al.

reproductive modes, with at least two different modes of key morphological characters inverted. Its classification as
reproduction occurring even within a single haplotype. These T. c. simplex is supported by the genetic data, very small telson
different reproductive modes may be associated with signi- length ratio (indicative of very short furcal spines), high number
ficant morphological differences, which in turn led to the of apodous abdominal segments, general appearance, geo-
assignment of populations that belong to the same haplotype graphical locality, and apparently equal sex ratio (53% males,
to two different subspecies (T. c. cancriformis and T. c. simplex). n = 17). Only the occurrence of numerous dorsal carina spines
A similar pattern has been reported in North American Triops: contradicts this classification, because T. c. simplex is diagnosed
the number of apodous abdominal segments in T. longicaudatus by complete lack of these spines (Ghigi 1921, 1924; Longhurst
was found to be 10.31 (SD = 0.755, n = 53) in females, but 1955). However, Gauthier (1934) described a specimen of
only 5.75 (SD = 0.483, n = 60) in unisexuals (see table 5 in Apus (= Triops) cancriformis ssp. simplex with five small spines,
Sassaman et al. 1997). These findings, together with our data, arranged in a regular row anteriorly, next to the terminal
suggest that in nongonochoric Triops, a reduced number of spine. This specimen was one of three females in a sample
apodous abdominal segments is linked to reproductive mode from Ghardaïa (Mzab, Algeria) that also contained one
rather than reflecting the specimens phylogenetic position. male. This was the only specimen of 73 from seven localities
reported by Gauthier (1934) as having dorsal carina spines.
The Triops cancriformis simplex group: no separate lineage Our investigation shows that the occurrence of dorsal carina
The only support for a separate T. c. simplex lineage comes spines in this taxon is much more frequent than previously
from the number of apodous segments in females, a character thought, even though the spines are usually very small.
that appears to be linked to reproductive mode in Triops spe- Longhurst (1955) may have accounted for this by calling its
cies and is thus of little value for phylogenetic classification of carina ‘rather smooth’. In our intermediate population, the spines
lineages that include nongonochoric populations (see above). appear to be of similar size to the spines of T. c. cancriformis
The key diagnostic character used hitherto to discriminate in which similar high numbers of spines have been observed.
the former T. c. simplex from other subspecies, i.e. the com- However, some Moroccan T. c. mauritanicus specimens with
plete loss of dorsal carina spines, proved to be erroneous: all reduced spines also have a rather similar appearance, although
gonochoric populations from which we were able to investi- the furcal spines of these Moroccan populations are much
gate a sufficient amount (more than three) of specimens and larger. Thus, morphologically the population in question
that were found in the distribution range of the former would be regarded as a T. c. cancriformis population with
T. c. simplex would have been determined as T. c. cancriformis extremely short furcal spines, and with an unusually high
using this character. number of apodous segments. Since genetically it is clearly
Our genetic data demonstrate that ‘Spanish T. c. simplex’ T. c. simplex, it could also be regarded as a cryptic lineage.
form part of the T. c. cancriformis lineage while ‘North African
T. c. simplex’ are part of the T. c. mauritanicus lineage. This is Taxonomic implications
supported by the obvious geographical isolation of both Our sequence data indicate a clear differentiation of
lineages (Fig. 2). Furthermore, geographical distribution of T. cancriformis into two main lineages (Fig. 4): T. c. cancriformis
both lineages suggests that ecological requirements are (including ‘T. c. simplex’ from Girona, northern Spain) and
basically different for these lineages: within Spain, the former T. c. mauritanicus (including ‘true’ T. c. simplex).
T. c. simplex appears to be refined to northern localities (Fig. 2), This result provides new insight into the relationships of
whereas the distribution of African populations of the former the subspecies of T. cancriformis. Mantovani et al. (2004) found
T. c. simplex suggests that they must be well adapted to desert the species to be very homogenous and thus ruled out the
conditions: they have, for instance, repeatedly been reported possibility of cryptic species within T. cancriformis. However,
in central Algeria (Gauthier 1934; Longhurst 1958). they did not include samples from within the geographical
Thus, we conclude that the T. c. simplex lineage as defined range of the subspecies T. c. mauritanicus and T. c. simplex.
in the most recent literature does not exist. They did include data from Japanese T. cancriformis, but these
appear to come from a unisexual population (Suno-Uchi et al.
The Triops mauritanicus lineage 1997) and this characteristic should have indicated the speci-
As currently constituted, T. c. mauritanicus is paraphyletic men to belong to the taxon T. c. cancriformis (referring to the
(Fig. 4), because the North African T. c. simplex samples nest data given by Longhurst 1955; on the reproductive mode of
within it. The southern Spanish samples of T. c. mauritanicus subspecies; see also Fig. 4). We therefore conclude that
form the sister group of the remainder of the lineage. they only examined one subspecies and could therefore
For the intermediate population from pond 063 (Kairouan) not make any conclusions regarding the entire species. Our
in Tunisia, comparison of genetic and morphological charac- study also reveals the T. c. cancriformis lineage as rather homo-
ters suggests that it is a T. c. simplex population with one of the geneous. The diversity of haplotypes is low with only four

314 Zoologica Scripta, 35, 4, July 2006, pp301–322 • © 2006 The Authors. Journal compilation © 2006 The Norwegian Academy of Science and Letters