In Reply.—In response to Dr Strickland, the locations of the wounds were described in the "Wounding" section of our report. As indicated, the wounds were in the paravertebral and thoracic areas. Unlike dogs and some other experimental animals, pigs have short, inflexible necks that prevent them from licking wounds on their own paravertebral and thoracic areas. When pigs are housed together, however, they are known to chew on each other's wounds. Therfore, we housed the experimental animals separately.
Dr Strickland was correct in pointing out that an analysis of variance of repeated measures should not have been used in these studies. We apparently did not clearly relate our experiments to those persons who advised us on the use of statistics. The data have been analyzed again using the logarithms of the counts, with a value of 0.30103 arbitrarily assigned in those instances in which no organisms were detected. Using