VOYAGE FROM NEW YORK TO RIO DE JANEIRO. 27 



yet wo have had no intensely warm weather. The suu, 

 however, keeps us within doors a great part of the day, 

 but in the evening we sit on the guards, watch the sunset 

 over the waters, and then the moonlight, and so while away 

 the time till nine or ten o'clock, when one by one the party 

 disperses. The sea has been so rough that we have not 

 been able to capture anything, but when we get into 

 smoother waters, our naturalists will be on the look out 

 for jelly-lish, argonautas, and the like. 



April 13^7i. In to-day's lecture Mr. Agassiz returned 

 again to the subject of geographical distribution and the 

 importance of localizing the collections with great precision. 



" As Rio de Janeiro is our starting-point, the water-system 

 in its immediate neighborhood will be as it were a school- 

 room for us during the first week of our Brazilian life. 

 We shall not find it so easy a matter as it seems to keep 

 our collections distinct in this region. The head-waters of 

 some of the rivers near Rio, flowing in opposite directions, 

 are in such close proximity that it will be difficult sometimes 

 to distinguish them. Outside of the coast range, to which the 

 Organ Mountains belong, are a number of short streams, little 

 rills, so to speak, emptying directly into the ocean. It will 

 be important to ascertain whether the same animals occur 

 in all these short water-courses. I think this will be found 

 to be the case, because it is so with corresponding small 

 rivers on our northern coast. There are little rivers along 

 the whole coast from Maine to New Jersey ; all these dis- 

 connected rivers contain a similar fauna. There is another 

 extensive range inland of the coast ridge, the Serra de 

 Mantiquera, sloping gently down to the ocean south of the 

 Rio Belmonte or Jequitinhonha. Rivers arising in this 

 range are more complex ; they have 2arge tributaries. 



