R. Pearl and F. J. Dünbar 329 



tho inner, :ur1 in i'iict is, so f'ar ;is absolute size ot' dmiation is conccrned. Taking 

 the size of tho doviations in relatiou to tho respcctive means, however, as has been 

 Seen, \vc find tho inner diaincter to be significantly moro varia])lc tlian tho outer. 



It is of intcrost to compare tho varial)ility of Aixella with that o( Faramoscium 

 as deterniined by Simpson*. In liis series of lOU Parariiaicia tho coefficient of 

 Variation for length was 8861 , and for breadtli l.'5-l'.'?!). 'J'hc vakie of this coeffi- 

 cient for length is less than either of the coefficients for tlic characters of Arcella 

 studied. The value of the coefficient of Variation for the breadth of Paramwcium 

 (13'439) is very nearly equal to that for the inner diameter of Arcella (13'6ö8). 

 Both the coefficients for Paramcecium are values of the same general order of 

 magnitudo as the coefficients for Arcella. From a comparative standpoint this 

 result is of considerable interest. It is highly desirable that quantitative studies 

 of the Variation in other Protozoa roproducing primarily by the niethod of fission 

 be made, in order to determine whether a vahie of ap})roxiniately 10 "j^ for the 

 coefficient of Variation is characteristic of Variation in individuals produced by 

 binary fission. The finding of such closely accordant vahies for the Variation in 

 two such widely ditferent protozoan groups as the rhizopods and the holotrichous 

 ciliates would seem to be of some significance. The senior author of this paper 

 hopes to carry on sonie further investigations bearing on this problom, u.sing other 

 species of Protozoa. 



It may again be noted in pa.ssing that in the sample of an Arcella population 

 here studied we havc a random general .sample of a iiopulation produced b}' 

 fission, the individuals not being chosen with reference to their innuediate ancestry, 

 as was tho caso in Simpson's work. It i.s intcresting to find .such close agreemeut 

 in the values for the Variation in .saniplos collected in these ditferent ways. 



It is of some interest to compare the size of the Arcellae found in this study of 

 a Single culture with the values given by Leidy from the results of his studies of 

 Arcellae from all parts of the country. Leidy saysf : " The shell of Arcella vulgaris 

 ranges from j^ij tu yij. of an inch in breadth, ^^^ to -^ of an inch in height, with 

 the mouth Trg'jpu to -g^ of an inch in breadth, and elevated ^-^^^^ to j^j of an inch.'' 

 Reducing the value.s for "breadth of shell" (outer diameter), and "breadth of 

 mouth " (inner diameter) to mikrons and comparing with our data, we get the 

 foUowing results. 



^ ., , , f Range of breadth of shell : ca. 49 — 1.52 mikrons. 

 Leidy s data < ■ i.i -7 <n -i 



*' ( „ ,, opening: ca. 127 — 49 mikrons. 



,-^ , f Range of outer diameter : ca. 42 — 9G mikrons. 



Our data 1 ^ /• ■ n <r> 



{ Range 01 nmer „ : ca. 11 — 40 „ 



(With only one individual e.xceeding 30 mikrons in inner 

 diameter.) 



It is Seen from this that the ränge observed by Leidy extends beyond ours only 

 at the ujjper end. 



* Biuineiriha, Vol. i. p. 405. t Loc. cit. p. 173. 



Biometrilca ii 42 



