2o8 SOLUiilLITY OF PHOSPHORIC OXIDi:. 



phosphoric oxide after shaking for half an hour in a shaking 

 machine. 



The quantity of material used, 4.366 grams, gave the fol- 

 lowing amounts of water-soluble phosphoric oxide: — 

 \'o. Material. Weight of Solu1)le Phosphoric Oxide. 



A. Pure Calcium Phosphate. 0.0164 gram. 



B. Bone A.sh. 0.0284 ,, 



C. Precipitated Tricalcium Phos- 



phate. 0.0793 „ 



Conclusions. 



Superphosphate can remain mixed for as long as three 

 weeks with either sulphate of ammonia or sulphate of potash or 

 kainit without an appreciable loss of water-soluble phosphoric 

 oxide, and if mixed with sulphate of ammonia there is a possi- 

 bility of an actual increase of water-soluble phosphoric oxide in 

 a period of three weeks. 



If immediate reversion of water-soluble phosphoric oxide is 

 to be avoided, Government guano should on no account be mixed 

 with superphosphate, for in a mixture of equal parts of the two 

 there is, even after three hours, a total loss of nearly 7 ])cr cent, 

 of the water-soluble phosphoric oxide. 



In the case of bone meal there is a loss of 2 per cent, of 

 water-soluble phosphoric oxide in three hours, but if left for 

 a period of 14 days there is considerable loss, amounting to 

 over 143/2 per cent. 



It is quite evident that the two samples of kainit as used 

 by Gray and myself varied considerably, judging from the loss 

 of water-soluble phosphoric oxide: the substance used by the 

 former cau.sed a loss of nearly 7 per cent, in 18 days, whereas 

 in my case there was a slight increase of water-soluble phos- 

 phoric oxide. These remarks apply also to the two bone pro- 

 ducts used : in the former case there was a loss of only 2>j4 

 per cent, in 18 days, as against 143/^ per cent, in 14 days with 

 the substance used by me. 



TRANSACTION'S OV SOCIIITIKS. 



SocTH .'\.FRic.\.\ I.NSTixrTiox OF Enginkkrs. — Saturday. -August 12th. 

 B. Price, M.I.E.E., President, in the cliair. — "Note on the value of ainieal- 

 itiji the connectinci attacluncnis on zvindvuj /'hints": J. A. Yaughan. 

 After receiving a heavy hlow from a skip to which it was attached, a 

 timber trailer working in an incline shaft on the l'"ast Rand broke loose 

 from tlie skip and dashed to the l)ottom of the shaft, killing si.x men who 

 were working l)elow. The direct cawse of the fatality was tlie doubk- 

 fracture of a ring connecting the traik^r with tiie skip. It appeared that 

 tin's ring had never been annealed. A portion of the fractured ring was 

 subsequently annealed, after which comparative tensile tests showed that 

 tile annealed portion possessed a greatly increased ductilit\ compared 

 with the unannealed portion; Iiad tliis ductibiiity l)cen present originally 

 it was possible that the accident migiit liave l)een prevented. — "./ few 

 notes on repairing a large vahr chamber by the qtiasi-arc system of 



