7 
marginals 2 of Mitrocystis . l I did, however, suggest in 1913 (par. No. 
80) that a hydropore would be found in Dendrocystis in the position 
where Messrs. Thomas and Ladd have now found an opening, and I 
actually described and figured a similar rounded projection in D. scotica 
and mentioned its occurrence in an unnamed species from H&rault (1913, 
par. Nos. 80, 136). This independent confirmation in D , sagittaria is, 
therefore, most gratifying. 
Enough has been said to show that this species falls into its true place 
as a Dendroqjstis, and that the resemblance to certain Anomalocystidae 
lies solely in the triangular outline, the exaggeration of the marginals, and 
the reduction in number of the body plates. This is a clear case of parallel 
evolution, but hardly close enough to be called convergence. It has no 
bearing on the relations between Dendrocystidae and Anomalocystidae, 
since it occurs in both at the end of the series not at the beginning. 
It will be noted that the larger somatic plates are on the reverse side, 
which, on my explanation of the mode of life, was that towards the sea- 
floor, whereas the obverse or uppermost side retains the smaller plates and 
the flexibility which they permit (Bather, 1913, par. No. 585). 
Dendrocystis (?) paradoxica Billings 
The kindness of Mr. E. M. Kindle and other friends enabled me, 
when visiting Ottawa in 1924, to study the original specimen of this 
species. It is from the Trenton limestone of Beaufort, Quebec, and is No. 
1521a in the National Museum of Canada at Ottawa. Thus something 
can be added to the remarks previously published by me (1913, par. Nos. 
155-159). 
Billings (1859) figured and described the fossil as showing portions 
of the stem alone. His figure shows parts of all three regions, but the 
specimen seen by me presents only the proximal and median regions. 
Some part of this specimen, as well as other specimens mentioned by him, 
has perhaps been mislaid. 
On the other hand, by cleaning the upper (adoral) part of the type 
specimen, I was able to expose some of the proximal plates of the theca. 
All those seen may be called somatic plates. They are thin and quite small; 
most are polygonal and irregular, but a few are hexagonal, with a diameter 
of 1-25 mm. On the left side of the specimen are one or two of the 
latter still in contact, but the other plates that remain are irregularly 
scattered, those between them having disappeared. 
The columnals of the proximal region are flanged rings formed of 
dimeres, and are very low or thin, as indicated in Billings’ figure, which 
represents the transverse ridges on the dimeres; but they are more num- 
erous than shown by Billings. 
The columnals of the median region differ in some respects from those 
of any other described species of Dendrocystis, and the points, which are 
not clear in Billings’ figure, are brought out in the accompanying sketch 
1 Bather: Palaeonfc. Zeitschr., VII, p, 13 (May, 1925). 
