310 PROCEEDINGS OF UNITED STATES NATIONAL MUSEUM. 



4. The Negretti-Zambra deep-sea tliermometers, which depend for 

 their self-registration upon the breaking of the mercurial column at a 

 certain place when the instrument is overset in pulling it up, have some- 

 times a trick of breaking the column in the wrong place, and so giving 

 a false indication. In one instance I noticed that the break was diago- 

 nal, instead of being directly horizontal, as it should have been. Pro- 

 fessor Hind, of Halifax, informs me that he has noticed the same defect 

 and has brought it to the notice of the makers, who have assured him 

 that it has been corrected in their more recent form of instrument. It 

 should also be always remembered that the temperature recorded by 

 these instruments is not that of the bottom, but of about a fathom 

 above it, owing to the play of line required in attaching them to the 

 sounding-line so that they may overset easily and not strike against 

 the lead. 



MODE OF OBSERVATION. 



The circumstances of the summer's work are too well known to you 

 to require repetition here. In explanation of the smaU number of 

 observations (ninety-seven for the whole summer) it will be sufficient to 

 refer to the unusual inclemency of the season, permitting not more than 

 an average of two excursions a week ; and to the remarkable scarcity 

 of fish, which made a large proportion of the excursions blank as to 

 results. Many fishes were brouglit up in the trawl-net of the Speedwell 

 (the naval steamer used by the Fish Commission), but had been so long 

 in the net, pressed upon by each others' weight, as to come up for the most 

 part dead; and always showing by their rectum temperature (which 

 should be near that of the bottom) that they were not in their normal 

 condition as to animal heat. Such observations as were taken from these 

 specimens are entered in the table (B), but arc not trustworthy for the 

 purposes of this investigation. On one occasion I set a trawl-line fur- 

 nished with some four hundred hooks, and took it in as soon as set. 

 Although not more than twenty minutes had elapsed between setting and 

 hauling, however, most of the fishes taken were already drowned, and all 

 had lost a large proportion of their animal heat. Since, therefore, no 

 tanks of sufiicient size for keeping fishes alive under observation were 

 available, there remained only line fishing, which was carried on during 

 the latter part of the summer as actively as the weather would permit, 

 from the yacht Phantom, belonging to the Engineer Corps of the United 

 States Army, and lent to tlie Fish Comuiission for the suuimer. The fish 

 were all taken in Cape Cod Bay, and witliin ten miles of Provincetown, 

 the two favorite localities being the steep edge of a shoal known as 

 "Shank-Painter Bar," between Wood End and Pace Point lights, and a 

 ledge in 15 fathoms of water some seven miles southwest of Wood End 

 light. 



The rectum temperatures indicate, and I have no reason to doubt, 

 that a fish caugiit with a line and hauled rapidly from the bottom to the 



