VON JHEEING : ON SOUTH AMERICAN MTTILID^. 85 



as My t this ovalis, Lam. (pi. xi, figs. 11, 12), belongs to Modiolus 

 purpuratus, Lam. Of this last - named species I have specimens 

 received from Dr. R. A. Philippi, while of Mytilus ovalis, Lam., I have 

 examples from St. Cruz, Patagonia. Here we have a Modiolus 

 described by a specialist as Mytilus, and vice versd. The same thing 

 has already been pointed out by Tapparone - Canefri, who, in his 

 "Viaggio della Magenta" (1873), p. 247, says that he has seen 

 the Patagonian Mytilus ovalis, Lam., in the British Museum as 

 M. purpuratus, Lam., which latter, however, is a Modiolus. On 

 comparing the two species, I observe that old shells of M. purpuratus 

 have the anterior extremity very stout and somewhat deformed, 

 and the anterior muscular impression very pronounced. There are 

 examples of M. purpuratus with and without teeth. The teeth are 

 \ also very variable in M. ovalis. Externally both are identical, and in 

 fact, I believe, both only local varieties of the same species. Morch 

 classifies M. ovalis with Modiolus. 



Mytilus falcatus, Orb., of Patagonia, described by me (Pevista Mus. 

 Paulista, torn, ii, p. 106) as Modiolus, is figured by Reeve both as 

 Mytilus and Modiolus ; his Modiolus strigatus (figs. 33 ' and 83) and 

 Mytilus sinuatus (fig. 16) are synonvmous, and, as I believe, identical 

 with M. falcatus. Orb., which has the same shape. Many of the 

 Argentine examples are ornamented with the same dark - green 

 markings as M. falcatus. 



It therefore seems necessary to unite these intermediate forms in one 

 s section, and, suppressing the section Brachydontes of Modiolus, to unite 

 5 all these radially striate species, having subterminal umbones, with 

 • Mytilus. 



'< Mytilus is said to be dimyarian, but there are both dimyarian and 

 ' monomyarian species. Among the Mytilus with smooth surface there 

 ! is a group of closely allied species which are all monomyarian. This 

 ; section comprises M. chorus, Molin., M. perna, L., and M. latus, 

 i Dillwyn. The absence of the anterior adductor is not the only 

 ! important character of this group ; there is also a distinctive disposition 

 > of the posterior retractor muscle. In the typical Mytilus, as also in 

 Modiolus, the posterior retractor has a straight insertion in the shell, 

 , forming, with the adductor scar, a figure ' 6.' In the Mytilus perna, 

 ' Linn., group, the posterior retractor scar is divided into two portions, 

 ' a separate rounded portion lying on the inner side of its pointed 

 - anterior end. This is the byssal muscle, which is well developed in 

 : M. perna, Linn., and M. latus. Lam., but in M. chorus, Molin., the 

 ' separation is sometimes incomplete, aud then the impression of the 

 i byssal muscle is contiguous to the rest of the posterior retractor scar, 

 i as seen, for example, in the figure of M. meridionalis, Krauss (Clessin, 



■ Mart.-Chemn., pi. viii, fig 8b). I have examined an animal of 

 i M. meridionalis, Krauss, in which both portions were quite separate, 



' as they are also in the Chilian M chorus, Molin. 



My observations on the muscle scars have been further completed 



■ by an examination of the rich collection of mussels, well preserved in 

 1 alcohol, in the Copenhagen Museum, that were placed at my disposal 



with great liberality in 1876. The want of the anterior adductor 



