Raymond Pearl and Wilbur N. Fuller 
215 
while it was alive, on account of the varying degrees of contraction of the muscles. 
This source of error is present to a certain degree in worms fixed and hardened in 
the way described above, but it is very greatly reduced in the fixed as compared 
with the living worm. Since, however, there was present in all cases a source 
of error on account of different degrees of contraction, no attempt was made 
to obtain close measurements of lengtl). What was desired was to get all 
the worms classified into a comparatively few, rather large length groups, for 
purposes of correlation. This end was attained by using as our smallest unit of 
measurement a distance of 2'5 centimetres. A worm was laid on a measuring 
stick divided into 2-5 cm. spaces, its length observed, and the record made by 
writing on the record sheet a letter with an algebraic sign, denoting the division of 
the measuring stick within which the end of the worm fell. Thus a record 
of (7 — indicated that that particular worm fell, in length, somewhere between 
10 cm. and 12-5 cm. In making the computations all worms falling within the 
class G — were considered to have been 11 '25 cm. long, or in other words it was 
assumed that within a group the lengths were arranged synmietrically about the 
centre of that group. The assumption is, of course, the same as is usually made 
in dealing with the frequency distribution of graduated variates. 
In the counts for the position and number of somites in the clitellum the 
following arbitrary rules were followed. The number of somites from the anterior 
end of the worm back to the first somite on which any clitellar tissue was present, 
was recorded as " Numbe]' of somites to clitellum." All somites on wJbich any 
clitellar tissue was present were recorded as " Number of somites in clitellum." 
Some such arbitrary rules were necessary in dealing with the cases in which the 
clitellum included only a part of one or both of its terminal somites. A record was 
kept, according to a plan which will be described later, of the exact extent and 
form of the clitellum in each particular worm. 
The only variations which were found in the collection of worms with reference 
to the position of the genital openings were regarded as teratological and the 
specimens exhibiting such variations were excluded from our tabulations. Abnor- 
mality in position of the genital pores in the very few cases in which it appeared 
was found associated with other abnormalities. The position of these openings is 
perfectly constant in normal worms. There is no " fluctuating " or continuous 
variation. On this account these characters will not be discussed further in this 
paper. 
It should be stated that the collection was made entirely at random. No 
selection on the basis of any characteristic of the worm, with the exception of the 
presence or absence of the clitellum as noted above, was exercised in forming the 
collection. The question of the homogeneity of the material will be considered 
farther on in the paper. 
We may turn now to the results. 
