DIRECTIONS FOR USING THE KEYS. 
229 
pancy is not beyond certain bounds. A variation of, say, five per cent, may be safely allowed 
on birds not larger than a robin : from this size up to that of a crow or hawk, ten per cent, j 
for larger birds even more. Some birds vary up to twenty or twenty-five per cent., in their 
total length at least. So if I say of a sparrow for instance, "length six inches," and the 
specimen is found to be anywhere between five and three-fourths and six and one-fourth, it 
will be quite near enough. But the relative proportions of the diflferent parts of a bird are 
much more constant, and here less discrepancy is allowable. Thus " tarsus longer than the 
middle toe," or the reverse, is often a matter of much less than a quarter of an inch ; and as it 
is upon just such nice points as this that a great many of the generic analyses rest, the neces- 
sity of the utmost accuracy in measuring, for the use of the keys, becomes obvious. When I 
find it necessary to use the qualification about" (as, bill a&ow^ = tarsus I probably never 
mean to indicate a difierence of more than five per cent, of the length of the part in question. 
It may be well to call attention to the fact, that most persons unaccustomed to handling 
birds are liable to be deceived in attempting to estimate a given dimension ; they generally 
make it out less than measurement shows it to be. This seems to be an optical eff'ect con- 
nected with the solidarity of the object, as is well illustrated in drawing plates of birds, which, 
when made exactly of life-size, always look larger than the original, on account of the flatness 
of the paper. The ruler or tape-line, therefore, should always be used, and particularly in 
those cases where analyses in the key rest upon dimensions. It is hardly necessary to add, 
that in taking, approximately, the total length from a prepared specimen, regard should be 
had for the ''make-up "of the skin. A little practice wiU enable one to determine pretty 
accurately how much a skin is stretched or shrunken, and to make the due allowance in either 
case. 
The measurements used in this work are all in English inches and decimals. 
There are probably no signs or abbreviations not self-explanatory or not already explained 
in " Field Ornithology." 
Fig. 112 his. — Diagram of corresponding segments of hind limbs of man, horse, and bird. The lines 1-11 are 
isotomes, cutting the limbs into morphologically equal parts, or isomeres. 
