528 



ADVENTURES IN RADIOISOTOPE RESEARCH 



than the corpuscles to which the plasma adheres. The activity of 1 gm of 

 the corpuscles secured from the active circulation is thus not strictly 

 comparable with the activity of 1 gm of the corpuscles injected. Knowing 

 the activity of the plasma at different times (cf. Table 2), we can correct 

 for this discrepancy. The red corpuscle value calculated without taking 

 regard to the above mentioned source of error will be 1.5%, 1.7% and 

 1.8%, respectively, too high. We could eliminate the above mentioned 

 sources of error due to the activity of the plasma by removing the active 

 plasma and replacing it by an inactive one. However, this step would 

 complicate the procedure. 



Not only, however, remains the error due to the adherence of the 

 plasma to the centrifuged corpuscles within the errors of the experi- 

 ment, it is In fact to a large extent compensated by the error due to 

 the penetration of active phosphate from the plasma into the corpuscles 

 during the few minutes which elapse between the injection of the active 

 blood and the collection of the samples. 



An estimate of the three errors of the experiment, — viz. (1) adhesion 

 of plasma to the centrifuged corpuscles, (2) penetration of ^sp from the 

 plasma into the corpuscles in the circulation of the rabbit (error due to 

 the fact that we do not inject labelled corpuscles but labelled blood), 

 (3) loss of 32p by the corpuscles during the experiment — is seen in 

 Table 3. After the lapse of 3 min., the plasma adhering to the corpuscles 

 is thus less active (per gm) than the corpuscles to which it adheres. Since 

 we determine the weight of the corpuscles + adhering plasma, and the 

 adhering plasma (per gm) is only l^ as active as the corpuscles, we find 

 the activity of the corpuscles too low, the error being 2—2 • 1/4 = 1.5 

 per cent. Correspondingly, after 5 and 7 min, respectively, we commit 

 an error of about 1.7 per cent. 



Table 3. — Estimate or Difeeeent Eeeoes of Expeeiment in the 

 Deteemination of the Red Coepuscle Volume 



Time in min 



Percentage error due 

 to adherence of the 

 plasma to the cor- 

 puscles 



3 



6 

 12 



+ 1.5 

 + 1.6 

 + 1.8 



Percentage error due 



to intrusion of ^"P 



of the plasma into 



the corpuscles 



-2.7 

 -3.4 

 -3.9 



Percentage error due 



to the loss of '*P by 



the corpuscles 



+ 0.5 



+ 1 

 + 1.5 



RESULTS 



As seen in Table 4, the corpuscle content of the rabbit per kgm of fresh 

 weight varied between 20.5 and 26.5 gm. It is difficult to decide whether 

 an erythrocyte reserve present in the spleen and some other organs 



