chanical or optical device for continuous recording 

 of changes in the position of the indicator. 



Both Hopkins' and Mironov's methods are of 

 hmited value because the mollusks can not be 

 kept in running sea water which would disturb the 

 recording devices. This greatly restricts the 

 application of their technique. 



Indirect methods 



All indirect methods are based on measurements 

 of the rate of removal of particles suspended in 

 water. Since the volume of water is kept con- 

 stant, the observations must be completed in a 

 relatively short time in order to avoid the effect of 

 metabolites. This condition seriously limits the 

 usefulness of the methods. 



Use of turbid water. — Viallanes (1892) was the 

 first to determine the relative rate of removal of 

 suspended particles by bivalves. He selected a 

 nimiber of small 0. edulis (18 months old), 

 C. angulata of the same age, and M. edulis of "an 

 average size" and placed them in separate crystal- 

 lizing dishes in a tank with running sea water. 

 Dishes of the same dimensions, but without mol- 

 lusks, served as controls. After several days the 

 sediment that accumulated on the bottom of the 

 dishes was collected, dried, and weighed. He 

 subtracted the quantity of material precipitated 

 mechanically in the controls from the total 

 quantity found in the dishes with mollusks, and 

 assumed that the remainder was proportional to 

 the volumes of water filtered by them. For each 

 liter of water filtered by 0. edulis, the C. arKjulata 

 filtered 5 1. and the mussel 3 1. Dry clay was 

 added to the experimental tanks in the proportion 

 of 0.0546 g./l. In 24 hours the mussel precipi- 

 tated 1.768 g. of clay, C. angulata 1.075 g. and 

 0. edulis 0.199 g. Essentiallj^ the same crude 

 method was employed 34 years later by Ranson 

 (1926). He did not record the temperature 

 during the obvervations and made no attempt to 

 observe the shell movements of the mollusks or 

 to note whether the valves remained open all the 

 time during the experiment. In later years the 

 ability of water-filtering bivalves to clear turbid 

 water has been studied by more elaborate methods. 

 The amount of material remaining in suspension 

 has been computed from turbidity observations 

 made by means of a nephelometer (Mironov, 1948) 

 or with an electrophotometer used as a turbidim- 

 eter (Lund, 1957). A great variety of suspen- 

 sions are used in this type of experiment — India 

 ink, colloidal graphite, carmine powder, powdered 



eggs, ground diatoms, fuller's clay, milk, calcium 

 carbonate, dried mud and others. Mironov 

 (1948) reports that the best material for this pur- 

 pose is nephelinic grey clay; after washing it 

 gives a very stable suspension in which the pre- 

 cipitation of particles is so slow that it has vir- 

 tually no effect on experimental results. 



In experimental studies of the rate of water 

 propulsion by the California mussel, Mytilus 

 calif ornianus, Fox, Sverdrup, and Cunningham 

 (1937) used a suspension of calcium carbonate 

 (CaCOa). The water was stirred continuously 

 to keep in suspension the calcium not removed 

 by the mussels. At frequent intervals analyses 

 were made of the calcium carbonate remaining 

 in suspension and from these data the rate of 

 water propulsion was computed. The rate of 

 precipitation of calcium in the control tanks 

 suggested that in the absence of mussels the 

 amount of calcium suspended in water can be 

 expressed as an exponential function of time and 

 that the amount precipitated in a unit of time is 

 proportional to the total amount which remains 

 in suspension. In the mathematical treatment 

 of the data Fox and his collaborators applied the 

 following exponential equation: 



P= 



nm 



=^»'^-(lf + 



a'\t=Poe-'" 



where p is millgrams of suspended calcium per 

 liter; Po is the amount of suspended calcium in the 

 closed sj^stem at the beginning of the experiment; 

 n the number of mussels in the vessel; m the 

 volume of water (in liters) transported by one 

 mussel in a unit of time; M the total volume of 

 water in the vessel; t the time; e the Napierian 

 base equal to 2.71828; and "a" and "b" the 

 logarithmic decremental constants determined 

 experimentally. 



Under the specified conditions of the experiment 

 in a closed system and using the above assump- 

 tions, the volume of water transported by one 

 mussel was calculated by the following equation: 



=M 



b—a 



The values obtained for medium sized mussels from 

 95 to 130 mm. long varied between 2.2 and 2.9 l./hr. 

 Several drawbacks are common to all methods 

 based on turbidity determinations. The inollusks 

 are kept in a closed system and are subject to ab- 

 normal conditions caused by high content of sus- 



TRANSPORT OF WATER BY THE GILLS AND RESPIRATION 



193 



