188 Thirty-fourth Annual Meeting. 
Mr. Clark: What would you call an inch and a half? 
Mr. Bower: I would call them No. 1. Above an inch and 
a half or perhaps two inches I would call No. 2. You will come 
a good deal closer to understand what is meant if you classify 
them according to length rather than according to age. As you 
have shown, starved fish may be only one-tenth the size of 
others of exactly the same age; but in classifying them by 
length you cannot be more than half an inch out of the way, 
either way. That is my idea of the proper method to grade 
the fish, so we may know very closely what is meant every time, 
whereas you cannot if you grade them according to age. 
Mr. Clark: As to the shipments per can, about which Mr. 
Lydell asked, he ships large mouth bass an inch long twenty 
days old, a thousand per can. This is about the same number 
we ship of the small mouth, except in cases of abnormal growth. 
President: I think Mr. Titcomb made the suggestion that 
they did not care much at the Fish Bureau what designation 
was given fish; they sent out what they had. Will you please 
state your mode of classification at the bureau? 
Mr. Titcomb: At the present time we have a plan of dis- 
tribution showing fry, fingerlings and yearlings. Of course that 
is very broad. Fry are the young fish until they have been fed 
for a time and are perhaps an inch or an inch and a half long. 
Then they begin to be fingerlings, and continue to be finger- 
lings until they are perhaps three inches long, which depends _ 
on the kind of fish; and then in the fall of the year, as they get 
larger, they are called yearlings. Mr. Bower suggests number- 
ing, as I suggested, only he has reference to the length of the 
fish, while I have reference to the age of the fish, and he has 
reference to one species of fish, while we want the committee 
to select it for all species. If we could, in our fish cultural par- 
lance, not only designate them as No. 1 for one month old, but 
in connection with that give the size or weight per thousana, 
we would know what we are talking about. But for generai 
tabular distribution, it seems to me that the numbering accord- 
ing to age would be suitable. In all this work I think we want 
to get into our tables and into our papers, not only a description 
