750 Transactions of section t 



the tissues, and does not contribute to the illness. The prevention is the arrange- 

 ment of the decompression period, so that the nitrogen dissolved in the body can 

 escape through the lungs without the formation of bubbles in the blood. 



The blood is the carrier by which the tissues are saturated and desaturated 

 with nitrogen. It is assumed that the blood is instantly saturated and desatu- 

 rated on passing through the lung3. The Admiralty tables of stage decompres- 

 sion are based on this assumption. 



Experiments by F. J. Twort, H. B. Walker, and the writer show that this is 

 incorrect. The subject stays an hour at +45 lb. Free diuresis is established 

 by drinking water. The bladder is emptied every seven minutes, after decom- 

 pression in five minutes from +45 lb. to +15 lb. Analyses of the urine show 

 equilibrium of the nitrogen after about fourteen minutes. It follows that the 

 arterial blood is only desaturated in about that time. 



The Admiralty tables and those of Mr. Japp (East River Tunnels, New York) 

 are based on the supposition that the blood circulates about once a minute ; on the 

 relative mass of blood and tissues ; on the relative solubility of nitrogen in blood 

 and tissues ; and on the supposition that ' quick ' parts half saturate in about five 

 minutes and ' slow ' parts in seventy-five minutes. 



The great difference between the circulation and pulmonary ventilation in 

 states of rest and hard work has not been taken into account. The circulation 

 may be six and even ten times faster during work. This upsets all calculations 

 based on ' quick ' and ' slow ' parts. Saturation and desaturation are not at the 

 same rate if carried out in the one case resting and in the other working. The 

 decompression period can be shortened by making men work during it. 



F. J. Twort and the writer have worked out the solubility of nitrogen in fat 

 at +90 lb., and find Vernon's figures for 1 atm. almost hold good for 7 atm. 

 Fat dissolves more than five times as much oxygen and nitrogen as water. These 

 experiments confirming those of Greenwood and the writer on heavy and light 

 rats, and these of Boycott and Damant on fatter and leaner guinea-pigs, show 

 that fat men are unsuitable for compressed-air work. 



It has been established by human experience that short exposures to high 

 pressures are free from risk. 



The Admiralty Committee, therefore, lays great stress on the danger of satura- 

 tion of ' slow ' parts, and says the danger increases with the length of shift. 

 While the effect of work may turn a ' slow ' into a ' quick ' part, the effect of 

 fatigue in long shifts complicates the problem. The writer regards the time 

 required for saturation and desaturation of the blood and abdominal organs, such 

 as the liver, as particularly the danger time. 



Neither the tables of the goat experiments of Boycott, Damant, and Haldane. 

 nor the tables of Keays concerning the 557,000 man-shifts at the East Kiver 

 Tunnels, give conclusive evidence that shifts of three hours are more dangerous 

 than one and a half, or eight than three. 



The variations in percentage of cases, even when calculated from groups of 

 3,000 to 4,000 man-shifts, are very large, e.g., eight-hours' shift 0'43 per cent. 

 May 1907, and 0'94 per cent. January 1907. Chance plays a very big role. 

 The first three-hour shift gave 035 per cent, cases, and nine fatal or dangerous, in 

 about 43,600 man-shifts; the second three-hour shift (after three hours' interval) 

 gave 072 per cent, cases, and four dangerous or fatal, in the came number of man- 

 shifts. The sum of cases for the six hours is 1'07 per cent. The percentage in 

 10,700 man-shifts of eight hours is 0"62. Two three-hour shifts with a three-hour 

 interval appears, then, to be almost doubly as risky as one eight-hour shift, 

 because it doubles the decompressions. The percentage of illness was 066, and 

 death - 0035, in 557,000 man-shifts, with a decompression rate of fifteen minutes 

 from +29 to +33 lb. 



Of the 3,692 cases among 10,000 men 89 per cent, were bends, 5 per cent, 

 vertigo = about 95 per cent, non-dangerous; 1'26 per cent, pain and prostration, 

 2-16 per cent, paralysis, 162 per cent, dyspnoea, - 46 per cent, collapse = about 

 5 per cent, dangerous. 



The compression relieved 90 per cent., and of the rest all except 0"5 per cent, 

 were partly relieved. The Admiralty table ordains a rate of 42 to 52 minutes 

 stage decompression for +29 to +33 lb. The writer does not expect engineers to 

 accept the Admiralty period in the face of the above figures. Eight thousand 



