KUNGL. SV. VET. AKADEMIENS HANDLINGAR. BAND 59. NIO 3. 67 



interlamellar septa are of alternating height and the afferent vessels are enclosed 

 in the higher, the efferent in the lovver septa. 



These facts agree with some statements foimd in literature as to the circula- 

 tion of the blood in the gills of Lamellibranchs. Ridewood's utterance (1903) about 

 the principal filaments may be quoted as comprehending all that is hitherto known 

 on this matter. Ridewood says (p. 162): 



»The differentiation of principal filaments may possibly be connected in some 

 obscure way with the fact that these alone contain the afferent branchial blood 

 channels, the arterialised blood finding its way back to the auricle from the ordi- 

 nary filaments. This direction of the blood stream is a matter upon which Bonnet, 

 Sluiter and Haren-Norman are agreed; but I am not in a position to express an 

 opinion upon it since, the whole of the material available being preserved in alcohol, 

 I have not injected any specimens. Ménégaux, it should be noted, regards the 

 blood as passing down one principal filament and back by the next, the principal 

 filaments of each lamella being thus alternately afferent and efferent (Solen and 

 some other forms), and Johnstone in his investigations of Cardium has arrived at 

 the same conclusion.» 



That the interlamellar septa alternate in height in most of the synaptorhabdic 

 Lamellibranch gills is a fact pointed out by Ridewood, as already mentioned. In 

 all probability it seems to be a general rule that the afferent vessels are contained 

 in the highest septa, the arterial ones in the lowest septa or, when all septa are 

 of equal height, into the apices of the interjacent plicae, as in Veneridae and some 

 other forms (ef. above). No comparative investigations on this matter have been 

 carried out so far, but this supposition is supported by all the facts known (ef., 

 for example, the conditions described by Vlés 1909 for Mya, Johnstone 1899 for 

 Cardium, as well as Ridewood 1903). 



If this supposition should pro ve to be correct, we have in the more or less 

 intimate coalescence between the principal filaments a good measure of the degree 

 of development and the mutual relation of the synaptorhabdic gills. The cases in 

 which the efferent vessels are situated in the apices of the plicae must be explained 

 — and this seems to be Ridewood's opinion too — as a more primitive condition, 

 because here no coalescence has yet taken place. When one begins, it seems to be 

 the afferent vessels that are in every case first subject to it. This phenomenon and 

 the formation of principal filaments and interlamellar septa seems to be explained 

 satisfactorily as follows: 



The branchial axis of Chamidae contains, as is shown by the figures, one 

 posterior (in other cases dorsal) afferent and one anterior (resp. ventral) efferent 

 blood trunk. This state of affairs may be general, at least in the higher Lamelli- 

 branchs. If we imagine a primary eleutherorhabdic stage with simple filaments in 

 which these vessels are contained, then, as a matter of course, they must enter into 

 the filaments in the same relative position : the afferent capillary thus gets its place 

 on the dorsal side of the efferent one. When these simple filaments become reflected, 

 which is carried out, as usual, by bending up in a dorsal direction, it results that 



