THE OSTRACODA. 221 



Comparable and Non-comparable Hauls. 



The comparable hauls of Ostracoda in the epiplankton were only 65. All hauls 

 ■with nets of 18 or 180 meshes to the linear inch were omitted : the former allow small 

 specimens to slip through ; as regards the very fine-meshed nets, it has become clear in 

 the course of the study of the Biscayan material that they are perfectly useless for a 

 proper representation of the macroplankton. presumably because the frictional resistance 

 of the meshes is so high, that such nets simply push the water in front of them aside 

 and allow very little to filter through them. A few more of the hauls have been rejected 

 because the material had been allowed to dry, because part of the catch had been 

 lost, &c. 



Of the serial hauls 36 e to 36 I, the deepest (36 I) has been completely worked out 

 for specimens over 1 mm. in length; of the other non-comparable hauls only 27a 

 (1250-0 fms.) has been completely worked out. There are limits to human patience, 

 and mine had been already reached when 36 d was finished. lu some cases, however, of 

 the rarer species, the non-comparable hauls have been gone over roughly in order to get 

 more specimens for measurement ; these specimens appear in the tables of lengths, but 

 not in those of distribution. 



Manipulation. 



The chief difficulties with an Ostracod are to hold the little bi-convex shells steadily for dissection, 

 and to get them exactly into a hoi-izoutal position for drawing and measurement. The tools needful are 

 a pair of mounted needles, the one blunt, the other as sharp as constant grinding will keep it; and a 

 microscope slide in which a convex hollow has been ground, such as is used as a " cell " for mounting 

 thick objects. 



The first thing is to get the specimen (if unfamiliar) into the right horizontal position for observing 

 and drawing its contour : it is put into the hollowed slide, and all superfluous fluid removed with a fine 

 glass pipette; this method is most useful, because the remaining film of fluid tends to hold the animal 

 steady by surface-tension. The animal is then pushed up the side of the cell, the curve of the hollow 

 being utilised to tilt it into exactly the right position with certainty. In the case of nearly spherical 

 forms, such as ylnhosa % , the animal can be supported by two fine threads of drawn glass in the 

 required position. The shape of the shell, which can now be accurately observed, yields on the whole 

 the most satisfactory specific character, and generally points to one of at most some half-dozen of 

 species. As far as possible, the contour should always be studied and drawn before dissection ; the 

 contour should never (if avoidable) be drawn from inside with the shells opened out flat, as this 

 procedure produces distortion near the hinge (dorsal border). The openings of the asymmetrical glands 

 should then be considered (the opening of the left gland of rotundaia can rarely be seen except from 

 the left side with the specimen strongly tilted ventrally). 



After this, it is generally necessary to open the shells. Again utilising the surface-tension of a thiu 

 film of water, the animal should be put on its back and held with the blunt needle, the sharp needle 

 inserted, first between the shells, and then between the left shell and the body, until it is possible to 

 make a cut with it through the occlusor muscles: the left shell can then be hinged back, cutting witii 

 the needle through the internal lining membrane if tough. Holding the animal with the blunt needle 

 about the middle of the body, cut with the sharp one round the base of the large muscle of the second 

 antenna; this appendage can then be hinged back, without being absolutely detached. All the 



38* 



