117 



specimens of a new epecies of that family. These specimens furnished a com- 

 plete embryological series, some of them having their eggs at the back of the 

 gills, between the upper pharyngeals and the branchial arches ; others their 

 young in their mouth in different stages of development, up to those a quarter 

 of an inch long and able to swim about, full of life and activity, when lemoved 

 from the g-lls and placed in water. The most advanced were always found 

 outside of the gills, within the cavity formed hy tie gill-covers and the wide 

 branchiostegal membrane. In examining these fishes Mr. Agassiz has found 

 that a special lobe of the brain, similar to those of the TrigJas, sends large 

 nerves to that part of the gills which protects the young, thus connecting the 

 cure of the offspring with the organ of intelligence." 



If we restrict the use of the word metamorphosis to designate those 

 changes which talre place in the ovum after deposition by the parent, whereby 

 we are presented with larval and nymphal forms, &c., we find amongst vertebrata 

 few instances of it. Of course the metamorphosis of the frog, toad, newt, 

 and Bairachia generally, has for ages befn known. No fish, however, was 

 supposed to go through a larval state until Auguste Miiller, in 1856, showed 

 that what had hitherto been supposed to be a particular species of fish was 

 merely the larval form of another well-known kind. I allude to the AmmocceUa 

 branchialis, the Sandprey or mudlamprey. This fish, eviilently one of the 

 Cyclostomes, and very similar in form to the common river Lamprey, differs 

 from it in the semi-circular form of the upper I'p and the absence of tfeth. 

 Miiller has shown that this supposed new genus of Cyclostomus fishes was really 

 the 3'oung or larval form of the Lamprey. When four years old the 

 edentulous and semi-circular mouth are exchanged for a circular multidental 

 mouth. It has been thought that the Leptocephali, as the Anglesea Morris, and 

 the Branchinstoma lanceolatum (Lancelet) of our own coasts were only incom- 

 plete larval forms of some known fish ; but from the recent researches of Dr. 

 Kowalesky and M. Bert, it would seem that the latter named fish is a fully 

 developed form. The eggs are said to be expelled by the opening of the mouth ! 



Ordinary sexual reproduction may be divided into DuBcioui and Her- 

 maphrodite ; of the former there are, as we have seen, three kinds, the 

 oviparous, ovo-viviparous, and the viviparous or placental. Of hermaphrodite 

 reproduction we meet with a great number of instances in the animal kingdom, 

 as in most of the molluscous and radiate classes, worms, some of the entozoa, 

 &c. No insects, no Crustacea exhibit normal hermaphroditism. Amongst the 

 Vertebrata naturalists long considered that normal Hermaphroditism never 

 occurred. I say normal, because abnormal instances of animals, as fish having 

 roe on one side and milt on the other, in the same individual, have long been 

 known. A few years ago, however, M. Dufosse, published an elaborate memoir 

 to show that a genus of percoid fish, the Serrantts, was normally bisexual, 

 and was able to fecundate its own ova. M. Dufosse's memoir is accompanied 

 by a plate shewing the anatomical arrangement of the generative organs, and 



