of Edinburgh, Session 1871-72. 
G61 
some more or less longitudinally directed axis. It is also subject, 
in various circumstances, to various revolutions or sinuous deflexions, 
in which its long axis moves through portions of curves which are 
measured by corresponding angles. On these curves and their 
influence I have made a few remarks while feeling deeply their 
imperfection and the need of much further observation and research. 
The student who has followed the argument in this paper will have 
observed the resort to inferences when direct observations would 
have been preferable. This remark applies to every subject dis¬ 
cussed in it; and while it is to be greatly regretted that such is the 
case, it is at the same time not to be forgotten that no method of 
making direct and exact observations has hitherto been discovered. 
The adoption of the homalographic method is surrounded with 
difficulties, not only in the method itself, but also in the procuring 
of subjects on which to use it; and while results obtained by it 
would be of great interest and importance, it is evident that they 
would not be complete or sufficient, for they can never be other than 
observations on parts in the repose of death, not in the turgescence 
and action of life. Until very recently, all our knowledge of the 
force of labour was on a like imperfect footing; but already ingenuity 
has suggested a means of basing this subject on exact observations, 
and Schatz has availed himself of these means, and greatly assisted 
us to arrive at results which we regard as probably the most impor¬ 
tant hitherto achieved in obstetric science. Till some ingenuity has 
succeeded in devising means of making like exact observations to 
settle the points discussed in this paper, we must be content to do 
our best to reach the truth by reasoning on what we do know more 
or less exactly. And it should be remembered that, by this method, 
we may reach the greatest assurance, if not certainty. A boy, play¬ 
ing with his dissected puzzle-map, may be certain that a county is 
rightly placed if it fits exactly into an entire hole formed of the 
conterminous boundaries of surrounding counties, especially if it 
also fits in nowhere else. So a theory which suits itself to all, or 
is in opposition to none, of numerous known conterminous condi¬ 
tions, may be, provisionally at least, assumed to be correct, and such 
assumption of correctness will vary with the number and testing 
character of the conditions so humoured by the theory. 
