214 Variation and Correlation in the Earthworm 
The worms were collected on rainy nights, in gardens and lawns near the 
campns of the University of Michigan. On such nights the worms leave their 
burrows, and it is comparatively easy, after some experience, to pick up a large 
number in a reasonably short time. 
The worms were brought to the laboratory, stupefied in weak alcohol, straight- 
ened to their full length, injected with weak chromic acid or formalin, and 
hardened and preserved in either alcohol or formalin. As the worms are stupefied 
in alcohol the muscles become relaxed, and it is easily possible to straighten them 
to their full length. They are hardened in this straightened position. The 
specimens used in this investigation were subsequently used by students for 
laboratory work. The measurements and counts were made during the first half 
of the fall semester of the college year, 1902-03. 
Quantitative determinations of the following characters were made on each 
worm : 
1. Total number of somites in the body. 
2. Length of body in centimetres. 
3. The number of somites from the anterior end of the worm to the most 
anterior somite of the clitellum. 
4. The number of somites included in the clitellum. 
5. The position of the genital openings. 
The following points regarding the methods used in making the determinations 
may be noted. In all counts of somites which began at the anterior end of the 
worm the prostomium was not counted, but the first somite which was complete on 
the ventral side of the worm was counted as 1. All the other somites in the body 
were included in the count. All worms having spiral segments or other segmental 
anomalies were excluded in this work. Such worms were preserved for use in 
connection with a general study of teratology in the earthworm. In making the 
counts it was found very helpful to use a " tally register," such as is used in 
counting telegraph poles, and by gate keepers at places of amusement for counting 
people. This instrument has been found very useful in collecting biometrical data 
on integral variates for a variety of characters (e.g., scale rows and gill-rakers in 
fish, etc.). Anyone who has ever attempted to make a series of counts of repeated 
like characters knows how difficult it becomes in a short time to " keep the place," 
both mentally and on the specimen. All this difficulty is overcome by the use of 
a tally register, and its use makes it possible to work for a long time without 
becoming unduly fatigued. The only thing necessary to secure both speed and 
accuracy is that the operator have some preliminary practice. 
With regard to the character " length of body " considerable care had to be 
exercised in order to get results in any way reliable. Of course, it would be nearly 
or quite impossible to get any reliable measure of the length of an earthworm 
