212 A Manual of Veterinary Physiology. 



these we may add a third, albumoses. If urine containing 

 globulin be saturated by prolonged shaking with crystalline 

 magnesium sulphate, a precipitate is produced which is 

 serum globulin ; the precipitate is best obtained with a 

 faintly acid urine. This test, however, applied to the urine 

 of the horse is liable to fallacy, for the reason that there 

 are certain other substances of a normal kind precipitable 

 by saturation with MgS0 4 ; the process, therefore, can only 

 be employed where albumin is proved by other tests to be 

 present, and when we are capable of recognising by sight 

 the difference in the physical character of a MgSO, albu- 

 minous precipitate and a non-albuminous one. Assuming 

 a proteid is thrown down by MgS0 4 the fluid is filtered, 

 and is then saturated by crystalline ammonium sulphate ; 

 this throws down serum albumin and albumoses, the former 

 in flocculent masses. By these tests we are capable of dis- 

 tinguishing the nature of the proteid excreted ; by the 

 following tests the presence of albumin only is indicated : 



(1) Take the reaction of the urine : if acid add a drop or 

 two of nitric acid and boil — a precipitate appears ; if the 

 urine be alkaline rather more acid will be required, but an 

 excess must be avoided, or acid albumin will be formed 

 which is not precipitable in an acid fluid on boiling. 

 (2) Boil the urine ; phosphatic turbidity may occur, or 

 proteid. if present, be precipitated. Add a drop of acid, 

 the phosphates or carbonates are at once cleared up, 

 whilst the proteid is unaffected. This test is an important 

 clinical one. 



When a urine to which nitric acid has given a turbidity 

 clears up on boiling and becomes turbid on cooling, albu- 

 moses — a peculiar variety of proteid — are present. These 

 appear to bo common in the horse. 



The proportion of albumin present in a urine is roughly 

 judged of by the height of the precipitate in the test-tube 

 after standing from 12 to 24 hours; thus we speak of '. or | 

 albumin; for clinical purposes this is often sufficiently 

 accurate. 



There are many other tests for proteid in urine, but 



