ANGLER FISHES GILL. 611 



empty egg-cases observed in the nest gave promise of an early opportunity of 

 seeing some embryos freeing themselves from their envelope. Meanwhile a 

 number of these eggs with live embryos were cut out of the nest and placed 

 in separate glass jars to multiply the chances of preserving them, while the 

 nest as a whole was secured in alcohol, as a memorial of our unexpected dis- 

 covery. The next day I found two embryos in one of my glass jars; they 

 occasionally moved in jerks, and then rested for a long while motionless upon 

 the bottom of the jar. On the third day I had over a dozen of these young 

 fishes in my raclj, the oldest of which began to be more active, and promised to 

 afford further opportunities for study. 



* * * But what Ixind of fish was this? About the time of hatching, the 

 fins of this class of animals differ too much from those of the adult, and the 

 general form exhibits too few peculiarities, to afford any clue to this problem. 

 I could only suppose that it would probably prove to be one of the pelagic species 

 of the Atlantic, and of these the most common are Exocoetus, Naucrates, Scopelus, 

 Chironectes, Syngnathus, Monacanthus, Tctraodon, and Diodon. Was there a 

 way to come nearer to a correct solution of my doubts? 



As I had in former years made a somewhat extensive study of the pigment 

 cells of the skin, in a variety of young fishes, I now resorted to this method to 

 identify my embryos. Happily we had on board several pelagic fishes alive, 

 which could afford means of comparison ; but unfortunately the steamer was 

 shaking too much and rolling too heavily for microscopic observation of even 

 moderately high powers. Nothing, however, should be left untried ; and the 

 very first comparison I made secured the desired result. The pigment cells of 

 a young Chironectes pictus proved identical with those of our little embryos. 



It thus stands as a well authenticated fact that the common pelagic Chiro- 

 nectes of the Atlantic (named Chironectes pictus by Cuvier) builds a nest 

 for its eggs in which the progeny is wrapped up with the materials of which 

 the nest itself is composed ; and as these materials are living gulf weed, the 

 fish cradle, rocking upon the deep ocean, is carried along as an undying arbor, 

 affording at the same time protection and afterwards food for its living freight. 



This marvelous story acquires additional interest if we now take into con- 

 sideration what are the characteristic peculiarities of the Chironectes. As its 

 name indicates, it has fins like hands; that is to say, the pectoral fins are 

 supported by a kind of prolonged, wristlike appendages, and the rays of the 

 ventrals are not unlike rude fingers. With these limbs these fishes have long 

 been known to attach themselves to seaweed, and rather to walk than to 

 swim in their natural element. But now that we have become acquainted 

 with their mode of reproduction, it may fairly be asked if the most important 

 use to which their peculiarly constructed fins are put is not probably in building 

 their nest. 



"While Agassiz soon reached the condusion that the nest maker was 

 the Antennariid so common in the seaweed, he did not continue his 

 investigation of the subject,'^ and it was many years before a more 

 searching inquiry into the constitution of the nest was made. Then 

 the subject was taken up by a French naturalist, Leon Vaillant. 



A number of the nests were obtained by Vaillant and the supposed 

 procedure of the nest maker was described in 1887.'' Only a summary 



"Agassiz died Dec. 14, 1873. 



^ Remarques sur la construction du nid de VAntennarius marmoratus Less, 

 et G., dans la Mer des Sargasses. CR. des seanc. de la Soc. de Biologie, (8) IV, 

 1887, p. 732-733. 



