122 PROCEEDINGS OF THE NATIONAL MUSEUM vol. iii 



Clypeal foveolae : Some taxonomists attach considerable importance 

 to the number of these supralabial pits, and even go so far as to 

 estabhsh species and genera on shght variations in the total number. 

 In some groups, such as the Rhinocricidae, the number may be fixed 

 and constant, in others, including the Atopetholidae, there is con- 

 siderable variation within a species. Our series of E. soleatus yielded 

 the following counts: 3-3 in 4 specunens, 3-4 in 1, 3-5 in 2, 4-4 in 4, 

 and 4-5 in 1, Cook found "five clypeal foveolae on each side, some- 

 times o\Aj four." Apparently the number ranges from 3-3 to 5-5, 

 with 4-4 occurring most frequentl}^ The foveolae appear to afford a 

 very weak basis for the establishment of any grade of taxonomic 

 category, particularly when it is based on a single specimen. 



Number of ocelli: Cook gave ocellus counts for one of his t3^pes, 

 43 on one side and 44 on the other. We have made counts from 4 

 specimens taken at random, and find considerable variation: 49-51; 

 49-45; 40-41; 44-46. So few numbers cannot be satisfactorily 

 analyzed statistically, yet it should be obvious that the number of 

 ocelli varies at least from 40 to 50 in each ocellarium, and the prob- 

 ability is that the expected range will be slightly greater. Here again 

 is another character which sliould be used w4th the greatest caution, 

 if at all, for the discrimination of species. 



It should be stated, moreover, that we anticipate the establishment 

 of geographic variational clines in many structural characters of 

 millipeds. If this proves to be the case, the total ranges of variation 

 within a species — particularly a widely distributed form — will doubt- 

 less prove to be much greater than we have indicated for material 

 from a single localit}^. 



Synonymy : Upon consideration of the known variability of Eurelus 

 soleatus, we can find nothing in the description of E. proximus to 

 warrant its recognition as a valid species or subspecies. The state- 

 ment was made that proximus is closely related to soleatus, but differs 

 in being slightly smaller, with 45 segments instead of 48 ("apparently 

 the normal and nearly constant number in soleatus^'), and in having 

 the first segment dark instead of light, as mentioned in Cook's descrip- 

 tion. Another point of difference drawn from the single type specimen 

 was that the posterior gonopod was extruded. 



The figures cited for both size and segment number in proximus are 

 included in the known range of variation in soleatus. The ostensible 

 difference in coloration is very unreliable inasmuch as the color of 

 preserved material is known to vary with both the method and duration 

 of preservation. The exposure of the posterior gonopod in the type 

 of proximus is of course meaningless, since these appendages are 

 extrusible in all spiroboloids; whether they happen to be protruded 

 at the time of preservation is entirely accidental, and does not so far 



