520 Transactions of the Society. 



4. SurirellcL 



In a very primitive form Eabenhorst, 1864 (22, p. 9, fig. 12 d), 

 gave a transverse section of Surirella. It is rectangular, with short 

 straight hnes at the corners. The sculpture of the valve was 

 minutely described by Pfitzer (19, pp. 108-10, pi. I., figs. 8-10, 

 pi. v.), and the former researches by Smith and Focke were con- 

 sidered. About the finer sculpture which produces the transverse 

 and longitudinal strise Pfitzer said nothing. The representation 

 of the coarser details may serve as a model, and I shall refer to it 

 frequently. The only species closely examined by me occurs in 

 fresh water, and I believe it to be 8. hiseriata, Ehrenb. The 

 minute drawing on the surface is like the well-known test-object 

 ;S^. gemma. From a single valve of this species I made a series of 

 transverse sections, from another a series of longitudinal sections, 

 and lastly a collodion cast of the inner side of a valve. 



The general outline of the valve, best seen from the cast, is 

 an elongated oval almost like a lancet. In the middle lengthways 

 is a ridge, on the margins are the wings as described by Pfitzer in 

 S. calcarata (19, pi. I., figs. 8-9, pi. V., fig. 6). The surface 

 between the middle line occupying the highest edge of the ridge 

 and the wings is bent somewhat wave-like. The wings are not 

 simple membranes, but are double, a fact already established by 

 Focke and Pfitzer. They are really folds in the cell-wall. Both 

 membranes adjoin closely in some places ; in certain intervening 

 spaces corresponding to the waves on the surface they are not 

 close together, but show a tube-like space. AU these communica- 

 tions end in a delicate continuous tube, which forms the tip of the 

 wing. 



The diagram of Surirella consists of (1) a midrib without 

 nodule or any other distinction ; (2) numerous transverse ribs which 

 extend at pretty regular distances from the midrib towards the 

 edges ; (3) transverse lines between these ribs and perfectly parallel 

 to them ; (4) longitudinal lines of extreme delicacy which cut the 

 transverse lines at right angles. We will now examine the result 

 of the transverse sections. The transverse section series com- 

 mences with a section the shape of which suggests that three or 

 four had been taken before ; the last, No. 66, has had at least ten 

 successors of equal thickness. From these facts we may appreciate 

 the delicacy and extreme usefulness of this series. Fifty times over 

 these sections substantiate the correctness of Pfitzer's images. I 

 have drawn Nos. 2, 9, 39, 40, and QQ. Of the longitudinal section 

 series I draw only the first, pi. IX., fig. 18, and another, fig. 19, 

 which goes through the middle of one valve, and is the most 

 instructive, as it shows most distinctly the waves on the sur- 

 face. My longitudinal section series has not that technical 



