OUTLINE OF DEVELOPMENT, CHRONOLOGY 67 
tions in the time of development in the oviduct and uterus, or 
to slow development before incubation in warm weather, or to 
individual variation. It becomes necessary, therefore, to find 
some other system. The method followed by a considerable 
number of investigators, namely to classify by the number of 
somites, has been found to be best between about the twentieth 
and ninety-sixth hours of incubation. In the table which follows, 
therefore, this method of classification is used. For the sake 
of brevity throughout the book a stage reckoned by the number 
of somites will be written 1s, 2s, 3s, etc. It is true that the rela- 
tive rate of the development of organs varies slightly. Never- 
theless, classification by number of somites is unquestionably 
the most exact method up to the end of the fourth day at least. 
Beyond this stage the method is difficult to apply, and after 
about the sixth day the number of somites becomes constant. 
After the fourth day the time of incubation is usually a suffi- 
ciently exact criterion for most purposes: the latent period has 
become a relatively inconsiderable fraction of the whole time 
of incubation, and the embryos that survive, assuming fresh eggs 
and normal temperature of incubation, are in about the same 
stage of development. 
Classification of embryos by length is a favorite method 
particularly in Germany, and it offers many advantages in the 
‘ase of some animals; under many conditions it is the only avail- 
able method. But it offers considerable difficulties, the most seri- 
ous of which come from the varying degrees of curvature of the 
embryo. In early stages of the chick, for instance, up to about 
12 s, the total length of the embryonic axis may be measured, 
for the embryo is approximately straight. The cranial flexure 
then begins to appear, and slowly increases to a right angle; 
during this period there may be an actual reduction in length 
of the embryo (cf. table, 14-16s). Conditions are also compli- 
‘ated by the fact that the head of the embryo is turning on its 
left side at the same time. The cervical flexure then appears 
and causes a second reduction of the total length (ef. table 29— 
32s). Later still the curvature of the trunk and particularly 
of the tail develops in somewhat varying degrees and makes 
bad matters worse. After these flexures are formed, let us say 
at about eighty hours in the chick, it is customary to take the 
so-called neck-tail measurement, that is, from the cervical flexure 
