1914] Sumner, et al.: Physical Conditions in San Francisco Bay 15 
a considerable number of hydrographic stations. But the Ekman 
sampler was found to have two rather fundamental defects, which 
rendered it inadequate for the purposes of the work in hand. 
In the first place, the stationary weight is placed altogether too 
high, so that the center of gravity is actually above the midpoint in 
the length of the instrument. As a consequence, the latter could not 
be depended upon to maintain its vertical position throughout the 
course of the descent. In the fairly swift currents of this bay it was 
found that the lower end of the tube began to veer as soon as it 
entered the water, and undoubtedly it struck the bottom at a consid- 
erable angle in most cases. 
In the second place, the pair of jaws which are intended to meet 
below the cutting edge of the tube, and prevent the escape of the mud, 
frequently failed to work, and the sample was partly or wholly lost. 
This was due to the very slight leverage exerted on these jaws by 
the pressure of the movable weight. The force was found to be 
searcely more than sufficient, even when the latter fell the entire 
length of the tube, and brought to bear a considerable impact upon 
the basal ends of the jaws. But commonly no great impact was 
exerted, owing to the tube’s being buried in the mud before the release 
of the weight. Thus the only force brought to bear on the jaws was 
oftentimes the mere pressure of the movable weight, after the with- 
drawal of the apparatus from the bottom. And it may be added that 
the clogging effects of the mud not infrequently prevented the ‘‘mov- 
able’’ weight from falling at all. 
A considerable number of samples were none the less taken with 
the Ekman instrument, though these were, on the average, much 
shorter than those taken with the apparatus soon to be described. 
At the suggestion of Professor Kofoid, pasteboard tubes were used 
in the Ekman sampler, certain minor modifications in the latter being 
necessary as a consequence. The pasteboard tube, with its enclosed 
mud, could be removed entire, and corked at both ends. This ex- 
tremely practicable idea was likewise adapted to the improved instru- 
ment which was later used. 
In view of the inadequacy of the Ekman sampling apparatus, and 
the importance of obtaining mud samples of the greatest possible 
depth, considerable thought was devoted to this matter by the scien- 
tifie staff of the ‘‘Albatross’’, and by Professor Kofoid and others. 
The problem was placed in the hands of Mr. Valdemar Arntzen, ex- 
pert mechanic of the University of California, who designed and 
