40 W. M. BALE ON SOME OF THE DISCOID DIATOMS. 



which contains a very robust variety, all three forms are 

 similarly associated. In the valves described as A. Janischii 

 the marginal sculpture differs somewhat from that proper to 

 A. splendens, but this is a necessary concomitant of the substi- 

 tution of primary for secondary areas. In A. splendens, as in 

 several other species, the secondary areas terminate in a sub- 

 marginal hyaline band, which encroaches slightly on the primary 

 areas at each side of it. When, however, all the areas have the 

 same structure, this band is wanting, all except the small portion 

 which properly belongs to the primary areas, so that a small 

 rounded hyaline patch opposite the edges of the compartments is 

 all that remains. 



The relationship between these forms has always appeared to 

 me obvious, as it evidently did to Ralfs, who describes A. splendens 

 as having a tooth on each compartment, or sometimes only on 

 alternate compartments. In order to obtain actual proof of this, 

 however, it occurred to me to examine some Peru guano 

 cleanings which had furnished numerous slides, but in which the 

 complete frustules of A, splendens, where they occurred, had been 

 left. I picked out ten of these and mounted them in balsam, 

 with the result that I found that three out of the ten contained 

 valves of the so-called A. Janischii, each being included in a 

 frustule between two of the normal valves. In all cases where I 

 have examined whole frustules of A. splendens I have found that 

 the two valves were either alike in the number of areas, or one 

 valve had a pair more than the other. Thus, if one valve had 

 sixteen areas it could be predicated that the other would have 

 fourteen, sixteen or eighteen. Where an internal disc was 

 found (A . Janischii) it had the same number of areas as one of 

 the outer valves. In the slide referred to one frustule had the 

 outer valves with fourteen and sixteen areas respectively, and the 

 internal disc with sixteen ; another had the outer valves with 

 sixteen and eighteen, and the inner with eighteen ; and the third 

 had twenty throughout. The areas of the inner disc have the 

 processes rather smaller than those of the outer valves, and 

 nearer the margin. Though the inner disc is usually smooth, 

 like the so-called A. glabratus, this is not invariably the case. I 

 have a specimen covered with reticulations as distinct as in the 

 typical valves. 



In Van Heurck's opinion several genera, as well as species, 

 have been founded on mere internal valves of various species of 

 Actinoptychus (as also of Asterolampra). Such are Debya and 



