THIRTEENTH ANNUAL REPORT 101 
well as in the adult specimens, and the heavily scaled parrot- 
fishes change as readily as the more lightly scaled red hinds or 
groupers. 
During the latter part of February, 1909, I had the oppor- 
tunity of seeing the fine collection of fishes living in the Ber- 
muda Aquarium, where many of the tanks are as large as those 
in the New York Aquarium. Observations on the colors of 
fishes made in Bermuda served to confirm in general those made 
in New York, but the results were not always the same, due, 
perhaps, to the fact that none of the tanks were supplied with 
white sand bottoms or rockwork linings. Some additional phases, 
however, were noted. 
The colors and markings are, to some degree, determined by 
the position of the fishes in the tank. When swimming in mid- 
tank the fishes are likely to have colors quite different from those 
assumed when resting against a dark background, while those 
displayed when at rest upon a bottom of white sand or gravel 
may differ from both. 
In the following notes I have frequently referred to the colored 
plates published in The Fishes of Porto Rico,* by Evermann and 
Marsh. In reply to my letter of inquiry respecting the methods 
used by the artists, Messrs Hudson and Baldwin, in making the 
original paintings for these plates, Dr. Evermann writes: ‘‘Mr. 
Baldwin’s work was done without an aquarium. Fresh, fre- 
quently living, specimens were put in his hands and he worked 
on them as rapidly as possible to get the colors before material 
change took place. 
“The paintings by Mr. Hudson were all made in Key West. 
The fishes were kept in a small portable aquarium. He was able 
to get specimens readily from the live-wells of the fishermen’s 
boats. 
“The Hawaiian paintings} (about which I had also inquired) 
were made from live fishes kept in portable aquariums. 
The Samoant fishes were painted from alcoholic specimens care- 
fully preserved, color sketches being made from life at the time 
the fishes were collected.” 
Dr. Evermann’s information as to the methods employed is im- 
portant in connection with the observations recorded in the pre- 
sent paper, as it enables us to define the particular phases of color 
in each species which the plates show. 
* Bulletin of the U. S. Fish Commission 1900. 
+ Bulletin of the U. S. Fish Commission, 1903. 
£ Bulletin of the U. S. Fish Commission, 1905. 
