Page:Transactions NZ Institute Volume 14.djvu/374

296 9. Scenedesmus quadricauda, R., xxxi.

obliquus, R., xxxi.

obtusus, R., xxxi.

I have included the Pediastreæ in this family for the sake of convenience, although the definition of the Desmids as given by both Hassall and Ralfs would exclude them, and Carpenter arranges them as a separate family. (When this paper was read I had not seen Mr. Maskell's article on the Desmidiaceæ of Canterbury. Since then I have considerably modified my remarks on this family, in order to avoid useless repetition and to bring it as far as possible into accordance with Mr. Maskell's paper).

Micrasterias ampullacea, var. (α) from Ruataniwha.

var. (β) Ruataniwha. This plant resembles var. β, in general outline, but differs in the smoothness of the edges of the fronds, in the flatness of the tips of the segments of the lobes, and in the absence of punctæ inside the margins of the frond. The flask-like shape of the segments is very marked. I have placed it under the head of M. ampullacea, var. β, for the present, but I am not sure that it will remain there permanently. Fig. 6.

Cosmarium cucumis? I have marked this with a note of interrogation, because I have seen only one specimen. It corresponds, however, so entirely with Ralfs' description in the smoothness of the frond, the deep constriction, and the equality of length and breadth, and the rotundity of the ends of the segments, that I have little doubt as to its identity.

Cosmarium tetraophthalmium is common.

margaritiferum abundant.

sp. n. Frond quadrate, slightly longer than broad, not deeply constricted. Segments conical truncated, united at the truncations, edges smooth. End view circular. This plant is remarkable for its peculiar hour-glass shape. I have found no description of anything like it in such books as have been accessible to me. If it prove a new species I should propose the name of C. clepshydra. Fig 7.

Staurastrum, sp. ? Edges of frond smooth, segments united below, divaricating. Upper and outer borders concave, uniting at an acute angle, inner slightly convex, lower rounded. On the upper concave border are three hyaline processes with smooth sides and flattened ends. At junction of outer and lower borders is one process similar to the upper ones but curved slightly outwards. This little plant is unlike anything in Ralfs—it might be an unusual form of S. lave, but in the latter the processes are forked, in the former not. Fig. 8.

Closterium. In this genus I have not found any specimen that could not be referred to Ralfs.