studies on marine Ostracods 119 



placed in the horizontal plane in order to bring about an optimal increase in the horizontal 

 projection; only one of them is in this plane — pointing almost straight forward — the other 

 points almost straight downward; moreover, both pairs are concentrated on the anterior side 

 of the shell. 



By a closer investigation of the methods of swimming in the different Ostracod groups 

 and by putting the results obtained in this inquiry into relation with the facts mentioned 

 above I think we are enabled to understand this problem. 



All Ostracods swim, at least as far as is known up till now, with the ventral side 

 downwards. With regard to the methods in which the limbs function in swimming three 

 different types can be distinguished: 



Type I: There is only one representative of this type, namely the peculiar species 

 Thaumatocypris ecMnata. The first antenna and the exopodite and endopodite of the second 

 antenna co-operate in swimming; they all carry out downward and backward natatory strokes: 

 in this the backward component is presumably the predominant one. (The species in question 

 is, at least as far as we know so far, a deep-sea form; up till now it has only been caught 

 once, with an open horizontal net at 1100 metres' depth. Observations as to its mode of 

 swimming have certainly not been carried out hitherto on living material; the information given 

 above is based exclusively on the structure and position of these limbs in pi. \n, fig. 3, G. W. 

 MtJLLER, 1906a; all the same I have very little suspicion of the correctness of this information). 

 Because of this in this genus, as in Bosmina, the body is pressed forward and upward in 

 swimming; in order to bring about a progressive motion in a straight line it is thus necessary 

 for other regulating factors to co-operate. 



Type II: To this type belong Cypridinids and Halocyprids (except 

 the genus Thaumatocypris). The first antenna and the endopodite of the second antenna do 

 not take part in swimming. Only the second antenna, which — apart from the endopodite — 

 has in these groups about the same structure as in the genus Thaumatocypris, functions in 

 swimming. Unlike what is the case in the latter genus, the exopodite of this limb does not 

 strike downwards and backwards in swimming, but outwards and backwards and somewhat 

 downwards. By means of this a progressive motion in a straight line is produced. Alteration 

 in the direction of motion is produced chiefly by increase and decrease of the force of the stroke 

 in the exopodite of one side or the other and by twisting of the joint between the protopodite 

 and the exopodite. 



Type III: To this type belong only the C y p r i d s*. In swimming the first antenna 

 strikes upward and backward and somewhat outward, the endopodite of the second antenna 



» A variant of this typo is found in tlic little family Polycopidae, which never has a pelagian life, as tar as 

 we know so far. (S. Lo Bianco mentions 1903. p. 152 a P o 1 y c o p i d in a plankton sample from a depth of 500 :ii. 

 1 can give no opinion about the value of this statement.) The only information that we possess as to the way in 

 which these forms move is found in G. W. MOh-er's work of 189'i, p. 15. According to this wTiler these animals lie 

 at the bottom for the most part. .,Hisweilen erheben sie sich in kurzen SprUngen vom Grund, schwimmen umher, 

 jedoch wenig anhaltend, entfernen sich aber nic weil vom Bo.len." I have myself observed a specimen of Polijcopc 

 setigera (see the special part of this work) living in an aquarium. The specimen was characterized by a straight-lined, 

 progressive, comparatively swift and ratlier tenacious method of swimming. If it was disturbed it rose from the 

 bottom and swam around with rapid natatory strokes for a long (sometimes several minutes) or short period, some- 



