40 



A JOl^NEY IN BRAZIL. 



will be difficult and often impossible in travelling, we must 

 at least get the nests of ants, bees, wasps, and the like, in 

 order to ascertain all we can respecting their communities. 

 When these are not too large it is easy to secure them by 

 slipping a bag over them, thus taking the whole settlement 

 captive. It may then be preserved by dipping into alcohol, 

 and examined at leisure, so as to ascertain the number and 

 nature of the individuals contained in it, and learn some- 

 thing at least of their habits. Nor le': us neglect the do- 

 mestic establishments of spiders. There is an immense 

 variety of spiders in South America, and a great differ- 

 ence in their webs. It would be well to preserve these on 

 sheets of paper, to make drawings of them, and examine 

 their threads microscopically." 



April 21st. — Yesterday Mr. Agassiz gave his closing • 

 lecture, knowing that to-day all would be occupied with 

 preparations for landing. He gave a little history of Steen- 

 strup and Sars, and showed the influence their embryologi- 

 cal investigations have had in reforming classification, and 

 also their direct bearing upon the question of the origin of 

 species. To these investigators science owes the discovery 

 of the so-called " alternate generations," in which the Hy- 

 droid, either by budding or by the breaking up of its own 

 body, gives rise to numerous jelly-fishes ; these lay eggs 

 which produce Hydroids again, and the Hydroids renew 

 the process as before.* 



" These results are but recently added to the annals of sci 



* As these investigations have been published with so much detail ( Steen 

 strap, Alternate Generation, Sars's Fauna Norwegica ; L. Agassiz, Contr. to 

 Nat. Hist, of U. S.), it has not been thought necessary to reproduce this par* 

 of the lecture here. Any one who cares to read a less technical account of 

 these investigations than those originally published, will find it in " Methods 

 of Study," by L. Agassiz. 



