816 SUMMARY OF CURRENT RESEARCHES RELATING TO 



taking up this position with regard to the relative truth of the respective 

 prints, I am asserting nothing but what is within my own knowledge ; 

 both Mr. Nelson and myself having produced exactly the same photo- 

 graphic image as Dr. Zeiss with the dry apochromatic 1/4 in. 



But to recur to the proofs of the compound structure of the Pleuro- 

 Sigma valve. It is not necessary that I should weary you by giving in 

 detail all the evidence I have collected, but will call your attention to 

 two prints only of a valve of one of the Pleurosigma I have taken at two 

 different planes. In both prints a bit of the valve is shown chipped 

 away, but while in the print taken at the lower level, the hole is clean 

 through, in the upper a fine grating is seen projecting over the hole, 

 and nothing, I think, can be more conclusive of different layers. Having 

 done my best to establish the existence of different layers in the 

 Pleurosigma valve beyond a doubt it now remains to determine, if 

 possible, the ultimate structure of each layer in one species, and then to 

 establish the nature of the ultimate structure as between one species 

 and another in the same genus. When a number of forms agree in 

 shape and their leading features, and the only difference between them 

 is the relative coarseness or the fineness of their structure, you cannot 

 draw a line and say, " Here ends truth and here begins error." It must 

 be true throughout or false throughout, and to establish the truth of the 

 one will establish the truth of the other. Let us see then ; first, what 

 are the leading features common to all the Pleurosigma, and secondly, 

 how far we can make sure of the ultimate structure of the coarsest form, 

 that is of Pleurosigma formosum,. It is not necessary to say anything 

 about the common shape which gives name to the genus, or the median 

 line, to an assembly like this, but I may mention one peculiarity of the 

 nodule common to all the species having the diagonal markings which 

 I do not think has been mentioned before. On one side 

 Fig. 108. of the valve there is simply a cavity at that point, but on 



^ v,^ the other side the median line at the nodule is joined as in 



f —4— fig. 108. My attention was first called to this by examining 

 a type-slide of the Pleurosigma, where I found it in form 

 after form. Another feature common to the same forms, 

 is two rows of perforations larger than the others running lengthways 

 on the valve, one on each side of the median line and two similar 

 rows, one on each outer margin. Lastly, in all the species having 

 diagonal markings there is the common feature of the structure being 

 composed of a square grating with a focal image formed in each 

 alternate square. This is enough, I think, to show that whatever the 

 structure may be it is of the same character throughout, and it now 

 remains by examples to find out if possible, what is the unit of that 

 structure. 



Whatever difference of opinion there may be about the truth of the 

 image of a structure when it recurs at regular intervals, there can be 

 none when you get an isolated particle or fibril, which, existing already 

 as a unit, cannot be the double or the quadruple of another unit. Such 

 a unit I have found of the structure of Pleurosigma formosum floated 

 entirely away from the valve. It seems to consist simply of a series of 

 short bars of silex placed lengthways on the valve, side by side, in such 

 a manner as to leave alternate interspaces between them (see fig. 109). 

 It will be seen from a study of this diagram how the ordinary appearance 

 of the Pleurosigma is produced. The larger interspaces, being produced 



