ANNALES DE L'INSTITUT OCÉANOGRAPHIQUE 3 



waters out to a depth of about 3oo meters; I have thus arrived at a kind of bottom-sampler, 

 two different kinds of which can be seen open fig. i. One of them has an opening of 

 0,2 m 2 and is made for digging deep into the sandy bottom, the other has an opening of 

 0,1 m 2 and is designed especially for digging in softer bottom. On fig. 2 they are both 

 seenclosed. When taken up from the bottom the contents of the bottom-samplerare sifted, 

 so that only the animals remain in the sieves while the bottom material filters through. 

 The individuals of each species are taken up with forceps, counted, weighed and the 

 result of a series of samples with these kinds of bottom-sampler from the same place in 



Fig. 1. — Two bottom-samplers opened. 



the sea are arranged on a table, which is intended to show how many animals are found 

 on 1 m 2 of the bottom (see fig. 3). The first column gives the number of individuals ( 1 ), 

 the next the rough weight of each species in gm. The total rough weight is given below 

 in gm. together with the quantity of organic dry matter in, not comprising the quantity 

 of lime from the shells of the animals. 



I may mention here, that o, 1 m 2 and 0,2 m 2 are very small areas in comparison 

 with the areas over which a dredge is generally dragged, but living animals have been 

 found in almost every sample taken with the bottom-sampler. Echinoderms generally 

 play an important role by their quantities; on large areas, for example, the bottom may 

 be covered with Amphiura, the arms of which form such a dense network over the 

 bottom that they may be compared to a fishing-net with small meshes spread out at the 

 bottom : each single small object, living or dead, falling down on these areas will in the 

 course of a few minutes be touched and examined by Amphiura and either eaten or found 

 unsuitable for food. 



(t) The Decimals in number of animals in this column in figs 3 and 9 are caused by division, as generally more 

 than 10 samples à 0,1 m' have been taken, for instance 20, 25 or 5o. 



