16 U. S. BUREAU OF FISHERIES. 



V. Birch bark, cold. — 20 kilos (44 pounds) birch bark is added to 1 barrel 

 water; to this is added 1 to 2 kilos (2.2 to 4.4 pounds) of soda. This 

 stands cold al)0ut 8 days to extract. The net lies in this 1 to 2 days. 

 VI. Linseed oil, raw. — The net is first artificially warmed ; it is then laid in 

 raw linseed oil for 1 day. It is then taken out, freed of oil, and dried 

 as in IV. 

 VII. Spruce cones. — To one-half barrel of water a quarter of cones is taken. 

 This is boiled vigoi'ously about 3 hours. After it is cooled ofl: the net 

 lies in it about 2 days. 

 VIII. Wood tar, Norwegian. — The net is artificially dried as in VI, laid under 

 warm wood tar for a sufficient time to assure that it is thoroughly 

 penetrated, after which it is taken out and freed of excess. 

 IX. Quebracho wood. — 10 kilos (22 pounds) to about 150 liters (40 gallons) 

 water, boiled for 1 hour, after which the net is laid in it while it is 

 still warm and allowed to lie in it a couple of days. 

 X. Zinc chloride. — 1 kilo (2.2 pounds) zinc chloride to 100 liters (30 gallons) 

 water ; net is allowed to lie in this 1 day. 



Before treatment a series of trials was made on different places in the skein 

 of thread used. The results of these stretching tests were as follows ; 



A. Hemp. 



B. Cotton. 



Average breaking load. 



Weight per meter 



Breiiking length 



14. 9 Mlos (av. of 13 tests). 

 0.623 gram. 

 2:5.9 kilometers. 



8.5 kilos (av. of 10 tests). 

 0.494 gram. 

 17.1 kilometers. 



It will be observed that in addition to the breaking load, Linde- 

 man also gives the weight per meter of the thread and " breaking 

 length." This latter figure denotes the length of thread that will 

 hold itself up, or the length of thread in a ball whose weight is the 

 breaking load. The strength of thread as ordinarily understood 

 means strength for a given size or weight, for of course a thread can 

 be made of any strength if it is only large enough. The smaller and 

 lighter a thread of given strength, the better. This figure, breaking 

 length, probably furnishes a better basis, if proper precautions are 

 taken, than breaking load, of the quality of a net. Since the weight 

 of a meter of thread is given and the breaking load is given, the 

 breaking load divided by the weight of a meter will give the break- 

 ing length in meters (1 kilometer=l,(^00 meters, or 0.6 mile). 



This breaking-length figure is particularly valuable in judging 

 such preservatives as tar and linseed oil, preservatives that add con- 

 siderable weight to the net. This figure may go to show that where 

 a net must support its own weight, one that appears to be well pre- 

 served may be, in fact, poorly preserved because of the greatly in- 

 creased weight. 



After testing the original thread, Lindeman treated different 

 samples by the methods described above, and put them, along with 

 untreated samples, in the sea, and after varying lengths of time meas- 

 ured strength and weight. 



The nets were put in the sea October 6 (1S95). 



First test, 8 days afterward. 



Second test, February 1 (1896), after about 14 weeks. 



Third test, ftlay 3, about 13 weeks after second test. After the third test the 

 samples were destroyed by a passing boat. A new series was prepared and 

 hung out in the sea July 7. 



Fourth test, October 1 (1S97), after V2i weeks. 



Fifth test, January 1 (1897), after 13 weeks. 



