Aveust 13, 1897.] 
all failures, the court-room experiment 
which declines to ‘go off’ is perhaps the 
most dismal. 
This brings to mind a kindred topic upon 
which there should be a word of caution ; 
laboratory experiments which work to per- 
fection may utterly fail when expanded to 
commercial proportions, so that it is wise 
to bear in mind the danger of swearing too 
positively as to what will happen in large 
plants, when the opinion is based only upon 
what is observed to occur upon the smaller 
seale. Like conditions will, of course, pro- 
duce like results, but it is marvellous how 
insidiously unlooked-for conditions will at 
times creep into one’s calculations, and how 
hard it is even to recognize their presence. 
When preparing his case for presentation, 
the expert often errs in not dwelling more 
largely upon certain points because he thinks 
them already old and well known. To him 
they may be old, but to the public they may 
be of the newest. Not only is the public 
unequally posted with the specialist, but 
What it once knew upon the subject may 
have been forgotten. Is is well, therefore, 
to insert, in ‘a special report, matters that 
would be properly omitted from a paper pre- 
pared for a professional audience. 
Sanitary problems are of especial interest 
to the public, but the amount of ignorance, 
or rather false knowledge, displayed con- 
cerning them is surprising and often diffi- 
cult to combat. The sanitarian is not un- 
frequently called upon suddenly to defend 
a position involving complex statistics; and, 
because the data cannot be forthwith pro- 
duced, the inference is drawn that his points 
are really without facts to support them 
and that they are consequently not well 
taken. 
Long before he gets into Court, particu- 
larly if the time for preparation of the case 
be short, the expert may well ‘ pray to be 
delivered from his friends.’ He may re- 
ceive a peremptory order by telegraph to 
SCIENCE. 
247 
‘determine the mineral qualities of this 
rock,’ when the telegram should have read 
‘ Assay this ore for silver;’ and later it may 
be a matter of surprise that a quantitative 
knowledge of the copper present was not 
obtained while passing along the line for 
the determination of the silver; for it is 
generally not known that the complete anal- 
ysis of anything is quite rare and corre- 
spondingly tedious and expensive. 
Toxicologists who hear me may call to 
mind some case involving a search for the 
presence of an alkaloid, strychnia for ex- 
ample, during which search the District 
Attorney, in his eagerness for information 
may have asked to know what the indica- 
tions were as to the presence of the poison, 
at a time when the extraneous organic 
matter was not nearly removed. He may 
have wished no final report, but only 
the simple probabilities, whereon to base a 
possible arrest. Such requests are very 
common, and are akin to a demand for a 
proof of the pudding during the early 
baking, when we all know that such proof 
comes at a much later stage of the proceed- 
ings. 
Finally, ‘““When doctors disagree who 
shall decide?” 
This question is often very vigorously 
settled by the jury, as was instanced in a 
recent celebrated murder trial in New 
York City. In that case what the experts 
had to say on either side was simply thrown 
overboard as a whole, and the finding was 
based upon the testimony of the remaining 
witnesses. 
What can be said upon this ques- 
tion of the disagreement of expert wit- 
nesses? First, it must be noted, they 
are far from being the only class of 
people who fail to agree, and that 
too on very important subjects. Do my 
hearers think it would be a very difficult 
task to find a small army of men who 
would testify very variously and very posi- 
