BY K. GKElG-SMlTll. 45 



amount of acid. In the duodenum and upper partsi of the 

 intestine of normal children the contents are slightly acid and 

 contain coccal forms almost to the exclusion of the rod forms. 

 Further and further down the tract, the hacilli become more and 

 more evident until, in the rectum, the cocci are comparatively 

 rare and the rod forms predominate (Tissier). This is rather 

 important, and is in agreement with the bacteriological analyses 

 of the healthy stools, in the majority of which no streptococci 

 were found. The fact that streptococci are more in evidence in 

 the stools of rachitic children shows that they persist for a 

 greater distance down the tract and that they are in relatively 

 theater numbers all along the canal. Their greater numbers 

 indicate a more acid condition of the intestinal contents, but it 

 is difficult to say whether they are the cause or the effect. Pro- 

 bably a vicious cycle has been set up. 



The rod bacteria thrive well upon sugar-free media, such as 

 nutrient agar, w^hile the streptococci grow better upon saccharine 

 media, and form verj' scanty growths in media devoid of sugar. 

 This appears to show that in cases of rickets in which strepto- 

 cocci predominate, the sugar derived from the food is in e.Kcess, 

 and the proteid is deficient in the intestinal contents. This is in 

 agreement with the experience of medical practitioneis who, in 

 treating cases of rickets, pre.scribe an alteration of the diet, 

 increasing the proteids and the fat, and diminishing the carbo- 

 hydrates. The effect of the increased proteid would be to 

 increase the relative number of the co^i-bacteria, while the 

 diminution of the carbohydrate would further accentuate the 

 difference by decreasing the streptococci. The increased fat 

 woidd supply the necessary energy and heat, and at the same 

 time offer a less favourable pabulum for the growth of the 

 streptococci. 



The case in favour of the streptococci being associated with 

 the disease is not, however, absolutely clear. Their occurrence 

 in the convalescent cases may weaken the argument. It is true 

 that in two of the stools they were of the salivary type, but the 

 third contained Str. iii., which was the typical race in the instance 



