5i6 



J. KAWARA. 



Races. 



All. 



L. F. 

 in m. 



W. F. 



in m. grams. 



W. D. 





5 



655 



146 



2.00 





aver. 







2 10 







690 



213 



2.76 







• 4S8 



186 



341 



Italian. 



3 



595 



191 



2.88 





aver. 



591 



197 



2.07 



It is a general belief among silk manufacturers, that the size 

 and weight of the filament of a cocoon is proportional to its weight 

 in deniers, that is to say, if a cocoon is larger and its filament 

 heavier, then its weight in deniers is greater, while on the contrary 

 if a cocoon is smaller and its filament lighter, then its weight in 

 deniers is smaller. But my researches show that this is by no 

 means always the case but that the contrary may happen. For 

 example, the filament of a cocoon of the races Awobiki, Akabiki, 

 Onichijira, etc. which measures more than 700 metres weighs less 

 in deniers than that of Koishimarii, Matamukashi, etc., which 

 measures less than 550 metres. In particular the filament of an 

 Awobiki cocoon weighs only ;is much as 1.93 Deniers. The 

 heaviest filament is, so far I have examined that of the Hakuriu 

 rcce, whose filament weighs 3.75 deniers. In view of the facts 

 detailed above, I do not hesitate to say, that the length of the 

 filament is not always proportional to the weight in deniers, and 

 similarly, the weight in deniers of a sinf^le filament is not always 

 proportional to the diameter of the same. The weight in dtniers 

 and the diameter of a filament of Hakuriu race, which are the 

 highest and largest of the filaments I have examined, are 2.1 1 and 

 30.9 respectively. But the filament of the Akabiki race which 

 weighs 2.32 denier measurs 30.4 n in diameter which is very nearly 

 equal to that of Hakuriu. P^urthermorc, although the denier of the 

 filament of Koishimaru is reckoned 2.46, which is lii^lier than 

 that of Akabiki, its diameter measures only 25.8 /i. Similar obser- 

 vations were also made in the Laboratoire de la ver a sole in Lyon 

 in 1885. Tlie results obtained in that laboratory were as follows : — 



