160 U. S. BUREAU OF FISHERIES 



upon the passage of light is not impeded as much as before. In stained 

 specimens, the dye colors the protein-bearing material (protoplasm) 

 within the lumen and renders it dark, especially on the sides where 

 the lumen is thickest. Owing to the thickness of the fiber, it is difficult 

 to render the structure of the fiber extremely clear, but in Figure 6 it 

 can be seen that the cotton hair placed in tap water for a short time 

 appears greatly different from the normal fiber. A cotton hair from 

 the sample placed in lake water which had been inoculated with slime 

 shows deterioration far advanced, while a similar sample in Dubos 

 medium is already badly disintegrated. One should notice that the 

 \\ndth of decomposing fibers is greater than that of unattacked hairs, 

 indicating again the destructive and disintegrating effect of the attack- 

 ing organisms. 



Several methods were tried for preparing fibers for examination. 

 The swelling test used by Thaysen and Fleming (1921) and coworkers 

 is altogether too delicate to use on fibers that have been extensively 

 damaged by long use or by tests in the laboratory. This test would 

 have a great value in testing the quality of cotton offered for sale in 

 the form of fishing gear. The Congo Red test of Bright (1926) is 

 useful for detecting the first degenerative changes, being less delicate 

 than the previous one; but this method is also too sensitive for the 

 preparation of material badly attacked and digested by the bacteria. 



The method finally adopted was simple. It was to mount and fix 

 a few fibers as for simple bacteriological staining and then to treat with 

 carbol-fuchsin for 20 to 30 seconds. Fibers which have suffered exten- 

 sive change are handled most satisfactorily by using a diluted carbol- 

 fuchsin and watching the progress of the staining. The cuticle does 

 not show clearly — a decided disadvantage since otherwise a picture of 

 the mode of entrance of the bacteria might be obtained. It does not 

 seem as if any method will be found to do this. It is probable that the 

 function of tars, incorporated in the various treatments used in the 

 bureau's experiment, has been to cover over the cuticle of the cotton 

 fiber with a substance resistant to bacterial attack, thereby delaying 

 invasion of the fiber by the organisms found in the water. 



When the tensile strength of the test threads began to decrease 

 rapidly, it was always found by microscopic examination that the 

 bacteria had penetrated the walls of the individual fibers and that the 

 two events were simultaneous. The fibers begin to swell after entry 

 is effected; this swelling is accompanied by a notable increase in the 

 number of the bacteria within the lumen and a wasting away of the 

 fiber from within until destruction is complete. 



Distribvtion of the bacteria and their environment. — In order to learn 

 the relative abundance and distribution of the bacteria a number of 

 samples of water were taken at fishing locations. These were obtained 

 by means of a bacteriological water sample apparatus, Wilson (1920). 

 With this apparatus sterile evacuated and sealed 8-inch test tubes 

 were lowered to a predetermined depth. A messenger was then sent 

 down to break the tip of a bent-glass tube which projected through 

 the rubber stopper. The broken tip allowed the entrance of water to 

 the container, which was then drawn up and taken to the laboratory 



