Raymond Pearl and Wilbur N. Fuller 
227 
variable number of somites but as a consequence of (a) any increase in the number 
of somites in this organ must necessarily entail a decrease in the number of somites 
between the anterior end of the worm and the anterior end of the clitellum. 
Hence a negative correlation between these two characters arises, (c) There is a 
tendency for the clitellum after being formed to increase its extent by growth. 
This growth tends to extend the clitellum both anteriorly and postei'iorly. In 
a sample of worms non-homogeneous with respect to age any extension in the 
posterior direction tends to increase the variation in the number of somites in the 
clitellum without correspondingly affecting the amount of variation in the number 
of somites lying in front of this organ, (d) The increase in extent of the clitellum 
by growth is more marked at the anterior than at the posterior end of the organ. 
If this analysis be correct the factors concerned in the form taken by the 
clitellum after it is first laid down are fairly simple and understandable. The 
problem as to what causes the clitellum to originally appear in the precise region 
where it does is not so simple however. It is a particular case falling under the 
general problem of the " localization of morphogenetic phenomena." While 
natural selection is capable of accounting for the facts when the worm is entirely 
normal, this explanation fails to account for such cases as those decribed by 
Bateson for Perionyx and referred to above, where the position of the clitellum 
bears the same relation to the other organs of the reproductive system, whatever 
its absolute position. 
4. The Taxonomic Position of the Form Studied. 
It was supposed that the sample of worms studied in connection with this work 
was sufficiently large to enable a very precise determination of the systematic 
position of the form to be made. The constants obtained for the various characters 
from this collection were compared with the values given by Beddard* for the 
same characters in the various species of Lumhricus. Somewhat to our surprise 
it was found that our specimens differed widely in some points from the condition 
found in any described species of the genus. The worms come nearest to L. hercu- 
leus, Sav., and should evidently be classed with, or at least very close to this species. 
The points of distinction and agi'eement are shown in the following table : 
Character 
Ann Arbor form 
L. herculeus Sav. 
(as given by Beddard) 
Number of Somites 
Mean 142-715 
180 
Length ... 
„ 191-7 mm. 
360 mm. 
Somites to Clitellum ... 
30-76 
31 
Somites in Clitellum ... 
6-24 
6 
(Other diagnostic characters in agreement in the two cases.) 
* A Monograph of the Order of Oligochaeta. Oxford, 1895, pp. 7'21 — 724. 
29—2 
