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



form, but in mud from Manila it has a broader outline, and 

 I found one valve perfectly circular. 



Actinoptychus. This genus is distinguished by its valves being 

 divided into six or more radial cuneate compartments, which are 

 alternately raised and depressed, the markings also differing (in 

 normal valves) on the elevated and depressed areas. On what 

 we may call, for want of a better term, the primary areas 

 {Hauptfelder of Schmidt), the coarse markings are usually more 

 robust, and often of different form, from those on the secondary 

 areas (Nebenf elder of Schmidt) ; further, the primary areas 

 usually bear a tooth or process near the margin, with, in some 

 species, a radial line connecting it with the umbilicus ; while the 

 secondary areas sometimes terminate in a submarginal hyaline 

 band, which is not found in the primaries. The fine striation 

 also is commonly different on the two sets of compartments. 

 The striation is generally fairly uniform within the limits of 

 a species, but the secondary markings, consisting of hexagonal 

 or irregular reticulation, or systems of branching veins, is most 

 variable in its distinctness, and is often wanting. When this 

 occurs it is generally assumed to be the result of the detachment 

 of the separate layer of the valve which is thus marked, but in 

 view of the fact that different valves exhibit every possible 

 degree of obsolescence of these markings, I have no doubt that 

 in many cases they have not been developed. 



Among the characteristics to which too much importance has 

 been attached in classification are the number of areas, the 

 substitution of primary for secondary areas (so that all the areas 

 are alike), the presence or absence of the secondary markings, 

 also of the lines connecting the umbilicus with the processes, and 

 the presence of small variations in the striation. The adoption 

 of these purely artificial distinctions has led not only to the 

 undue multiplication of specific names, but, what is worse, to the 

 lumping together of forms which are by no means closely related. 

 In several species there are six areas, a number which is 

 rarely, if ever, departed from. Such are the forms composing 

 the group of which A. boliviensis is typical. In the majority of 

 species there is no constant number; for example the beautiful 

 A. Heliopelta, valves of which usually have six, eight, ten, or 

 twelve areas (constituting Ehrenberg's four species of Heliopelta), 

 while more rarely there are fourteen or sixteen. A. undulatus, 



