56 



Mr McConnel, On Measurement 



[Nov. 26, 



between them. These however, since the points were so far apart, 

 were still more subject to the error I have described. Indeed it 

 would probably amount to 10' in the second ring and 5' in the 

 third. Taking this into consideration, they agree fairly well with 

 the former observations. But as it is rather a haphazard correc- 



ts 



3 



