570 



NATURE 



[October 26, 191 1 



until the partial pressure ol oxygen dissolved in the blood 

 ties such a point that it acts as a tissue poison ; (4) that 

 the illness which occurs on decompression is prevented by 

 making the period of decompression sufficiently slow, by 

 allowing time for the dissolved nitrogen to escape from 

 the lungs. Looking through the works of Robert Boyle, I 

 found that, after the invention of his air-pump, he " had 

 a mind to observe whether when the air from time to time 

 was drawn away, there would not appear some hidden 

 swelling, greater or less, of the body of the animal by the 

 spring and expansion of some air (or aerial matter) in- 

 cluded in the thorax or the abdomen." He recorded that 

 a viper's body and neck grew prodigiously tumid ; that a 

 bubble of air appeared in the aqueous humour of a viper's 

 eye ; that the heart of an eel grew very tumid and sent 

 forth little bubbles ; that blood boiled " over the pot " until 

 the blood occupied only one-quarter of the volume of the 

 whole, so great was the expansion of the bubbles given 

 off from it. In the following surmise, concerning the death 

 of animals submitted to rarefaction, Boyle forestalls Bert. 

 " Another suspicion we should have entertained concerning 

 the death of animals, namely, that upon the sudden re- 

 moval of the wonted pressure of the ambient air, the 

 warm blood of those animals was brought to an efferves- 

 cence or ebullition, or at least so vehementlv expanded as 

 to disturb the circulation of the blood, and so disorder the 

 whole economy of the body." 



Hoppe-Seyler (1857) demonstrated bubbles in the blood- 

 vessels of animals submitted to rarefaction. This was 

 denied by Bert, but confirmed in the case of a rabbit bv 

 Greenwood and myself. 



Out of thirty autopsies done on fatal cases of caisson 

 illness, in nineteen gas-bubbles were visible in the blood- 

 vessels ; of the other cases most were old-standing lesions 

 of the spinal cord. 



The paralysis so often produced is due to a local death 

 and degeneration of the spinal cord, produced bv bubbles 

 blocking the circulation there (v. Schrotter, Heller, and 

 Mager). 



Proofs that nitrogen gas dissolved in the bodv fluids and 

 fat is the cause of the illness are the following. The 

 blood collected from the artery of an animal while under 

 pressure, and analysed with the gas-pump, shows that the 

 amount of dissolved nitrogen varies with the pressure. 

 Roughly. 1 per cent, per atmosphere is dissolved (Bert, 

 Hill, and Macleod). 



Exposed to one atmosphere at bodv temperature, blood 

 dissolves just about 1 per cent. N, to two atmospheres 

 2 percent., to three atmospheres 3 per cent., and so on. 

 The tissue fluids take up the dissolved gas from the blood, 

 and with time the whole body becomes saturated, accord- 

 ing to Dalton's law. The saturation of the bodv fluid- 

 takes time, since the blood forms but 5 per cent, of the 

 whole body weight, and it is the blood alone that comes in 

 direct contact in the lungs with the increased atmospheric 

 pressure. Probably about 5 kilograms of blood circulate 

 through the lungs per minute, and this blood convevs the 

 absorbed nitrogen to the fio kilograms of tissues.' The 

 arterial Wood saturated in the lungs yields the nitrogen to 

 the tissues, and returns to be saturated again in the lungs 

 Those tissues which are plentifully supplied with blood 

 will become saturated rapidly, while less vascular areas, 

 and parts in a state of vasco-constriction, will saturate 

 verv slowly. 



•n'j Hrlm and T ex P° setJ rats t0 ten t0 twenty atmospheres, 

 killed them by instant decompression, and then, opening 

 their bodies under water, collected and analvsed the gas set 

 free therein. We obtained in this gas CO, 6.7 to 16 per 

 cent. O, 2-1 to 8-7 per cent., N 80 to 87 per rent., and 

 a volume of N greater than that calculated according to 

 solubility of N in tissue fluid. Some of the excess we 

 found was due to air swallowed while under pressure the 

 rest to solution of N in fat. 



M. Greenwood and I have tested upon ourselves the 

 rate of saturation, using the urine ns a test fluid. We 

 were compressed in a large boiler, placed at our disposal 

 by Messrs. Siebe, Gorman and Co. The chamber was 

 fitted with electric light and telephone, and taps for slow 

 decompression. The pressure was raised bv means of n 

 divmg-pump driven bv a gas engine. We drank a quart 

 of water before entering, and eollerled samples of urine 

 NO. 2 191, VOL. 87] 



at varying pressures and times. The urine, collected in 

 sealed bulbs, was evacuated by the blood gas-pump. We 

 found the urine secreted in the next ten minutes after 

 reaching any given pressure is saturated with N at that 

 pressure. 



To demonstrate the bubbling off of nitrogen on rapid 

 decompression, I have spread the web of a frog's foot or 

 wing of a bat over the glass window of a pressure chamber. 

 The circulation of the blood is projected on a screen with 

 aid of microscope and arc light. We can thus observe 

 the circulation under twenty atmospheres of air, and watch 

 the bubbles forming in the capillaries on rapid decom- 

 pression. Recompression diminishes the size, and finally 

 drives the bubbles again into solution. 



When the larger mammals are exposed to high pressure, 

 such as eight atmospheres, for an hour or so, and are then 

 rapidly decompressed, they usually die in a few minutes. 

 Small mammals, such as mice and rats, may escape, owing 

 to the small bulk of body and rapid respiration and circu- 

 lation. The young of rabbits, cats, &c, also escape more 

 frequently than old animals. This is due rather to their 

 smaller weight and more rapid circulation than to the 

 youth of the body tissues. Paralysis in the limbs follows 

 too rapid decompression, or the animals fall over and 

 become unconscious. Noise of gas bubbles gurgling in the 

 heart may be heard. Respiration becomes embarrassed, 

 and the animals die. On dissection, the peritoneal cavity 

 may be found distended with gas, or the stomach, and gas 

 may be seen in the intestine. A part of this gas arises 

 from the fermentative processes of digestion, and from air 

 swallowed during compression. The veins of the portal 

 system, the venae cavae, are seen to contain chains of 

 bubbles ; the heart is full of froth. Small haemorrhages 

 may be present in the lungs. The edges of the lobes of 

 the lung are emphysematous, blown out by the rapid 

 decompression. The fat often is full of small bubbles. 

 so too are the connective tissues. Bubbles are seen in 

 the joints, and may appear in the aqueous humour of the 

 eye. On opening the skull, bubbles are seen in the wins 

 of the brain. The bubbles are not restricted to the veins, 

 but may also be seen in the arteries. The coronary vessels 

 of the heart often show chains of bubbles. On micro- 

 scopic examination the bubbles are seen in the capillaries : 

 here and there they run together and form larger bubbles, 

 sometimes rupturing the walls of the vessel and compress- 

 ing the surrounding tissues. In the larger animals, decom- 

 pressed from 100 lb. in four to seven seconds, we have 

 found the cells of the liver, kidney, &c, vacuolated or 

 even burst by bubbles. The gas set free in the heart can 

 be collected and analysed; about 80 per cent, of it is found 

 to be nitrogen (Bert, v. Schrotter, Hill and Macleod). 

 Catsaras lowered dogs in a diving dress to depths of 

 4.V7 m -, and after about an hour rapidly drew them to 

 the surface. He found bubbles set free in these dogs just 

 as in those exposed in a pressure chamber. 



In animals which escape without any severe symptoms, 

 some gas bubbles may be found in the veins even six hours 

 later. This shows how long it may take for nitrogen gas 

 once set free as bubbles to escape from the lungs, and 

 explains why caisson workers mav suddenly be seized some 

 half-hour or more after leaving the works. In such cases 

 the bubbles may be swept from the abdominal veins — 

 where they do no harm — into the heart, and impede the 

 action of this organ, or they may penetrate the pulmonary 

 circulation and enter the arterial system, and block up. 

 perchance, the coronary arteries, or others in the brain 

 or spinal cord. 



The blood is a colloidal solution, and it takes time for 

 the nitrogen to come out of solution and for the small 

 bubbles to run together to form visible bubbles. The gas 

 bubbles tend to collect in the veins, as the blood travels 

 quickly through the arteries and slowly in the veins. It 

 is only when the gas in the veins becomes sufficient in 

 amount to produce foam in the heart, or when gas bubbles 

 block ur> arteries of vital import, that grave symptoms 

 arise. The place where bubbles in the arteries must 

 always produce serious results is the central nervous 

 system. In the liver, kidneys, muscles, fat. &c. bubbles 

 may embolise small arteries and produce no grave effect. 

 but in the spinal 1 ord the interruption of the blood suoplv 

 to any group of cells or tract of fibres is evidenced at 



