VOYAGE FROM NEW YORK TO RIO DE JANEIRO. 35 



ally of low sand banks ; but this morning we have sailed 

 past the pretty little town _qf Olinda, with its convent on 

 the hill, and the larger jcity of Pernambuco, whose white 

 houses come quite down to the sea-shore. Immediately in 

 front of the town lies the reef, which runs southward along 

 the coast for a hundred miles and more, enclosing between 

 itself and the shore a strip of quiet waters, forming admi 

 rable anchorage for small shipping. Before Pernambuco 

 this channel is quite deep, and directly in front of the town 

 there is a break in the reef forming a natural gateway 

 through which large vessels can enter. We have now left 

 the town behind, but the shore is still in sight ; a flat coast 

 rising into low hills behind, and here and there dotted with 

 villages and fishing-huts. 



The lecture on Saturday was rather practical than scien 

 tific, on the best modes of collecting and preserving speci 

 mens, the instruments to be used, (fee. To-day it was upon 

 the classification of fishes as illustrated by embryology ; the 

 same method of study as that explained the other day and 

 now applied to the class of fishes. &quot; All fishes at the time 

 when the germ becomes distinct above the yolk have a 

 continuous lin over the whole back, around the tail, and 

 under the abdomen. The naked reptiles, those which have 

 no scales, such as frogs, toads, salamanders, and the like, 

 share in this embryological feature of the fishes. From this 

 identity of development I believe the naked reptiles to be 

 structurally nearer to the true fishes than to the scaly rep 

 tiles. All fishes, and indeed all Vertebrates, even the high 

 est, have, at this early period, fissures in the side of the neck. 

 These are the first indications of gills, an organ the basis for 

 which exists in all Vertebrates at a certain period of their 

 life, but is fully developed and functionally active only in 



