154 DISCUSSION 



therefore on the time after injection that the assay is made. I would 

 suggest the desirabihty of some independent test of immunity or 

 tolerance beyond the criterion of spleen enlargement. 



A second point is that the Friend virus, which is carried in many 

 strains of mice, has as its chief manifestation the production of splenic 

 enlargement on transfer to neonatal animals. This might be an addi- 

 tional, though one hopes unnecessary, reason why one would desire an 

 independent assay. 



Michie: On the question of desirability of independent tests of 

 immunity, we entirely agree, but mice and time are not all that plenti- 

 ful. We have started practising skin grafting to baby mice. 



About the variability of spleen enlargement: everything depends, 

 first on littermate control designs, by which you can cut down the 

 variability by a very large factor, and secondly on the nature of your 

 built-in controls. If the design is as it should be, it is possible to estimate 

 what the error variation really is, from whatever cause. If there are 

 viruses around, then they will be just as likely to manifest themselves 

 in the untreated member of a litter or, say, an isologous-injected mem- 

 ber of a litter. When you speak of the great variability which you have 

 experienced are you confining attention to within-litter variation 

 assessed in young hybrids of, say lo or 12 days old? This is the age at 

 which we kill them and weigh their spleens. 



Hildemann: Yes. We have also used littermate controls and have 

 given careful attention to the design you have indicated. 



Brent: I don't quite see why the transfer of sensitivity with presen- 

 sitized lymph node cells should necessarily rule out the hypothesis of 

 enhancement. 



Michie: The method has been used to exclude a category of unres- 

 ponsiveness under the general heading of peripheral block and efferent 

 block. 



Brent: This would apply only to unsensitized, normal cells rather 

 than to presensitized cells. The presence of antibody does not affect 

 presensitized cells. 



Michie: It is not sufficient protection? 



Brent: No, it is not. Your experiment would have to be done with 

 normal, unsensitized cells, to permit you to distinguish between these 

 two possibihties. 



