ALTITUDES. 251 



Next, the very accurate surveys of the Northern Railroad between 

 Concord and White River Junction, made by A. M. Shaw, were accepted 

 as correct, the difference between the extremes of the road being 1 16.84O 

 feet The height of track at White River Junction is thus placed at 

 369.237. The accuracy of these levels between Concord and White 

 River Junction is confirmed by the levels of R. S. Howe, engineer of 

 the Concord & Claremont Railroad, which differ from those over the 

 Northern Railroad by only a small fraction of a foot. 



To obtain the remainder of this base line, recourse was had to a special 

 survey under A. F. Reed, in 1871, assisted by Dr. N. Barrows of Mer- 

 iden, and Messrs. C. F. and F. A. Bradley of Dartmouth college. This 

 series extends over the Connecticut & Passumpsic Railroad to Barnet, 

 Vt, at which point it leaves the railroad, and follows the carriage-road to 

 Connecticut lake. 



A line of levels from the sea, connecting with this series at Dalton, is 

 furnished by the very reliable surveys for the Portland & Ogdensburg 

 Railroad, crossing New Hampshire through the heart of the White 

 Mountains. The connection between these series was made for the geo- 

 logical survey by James T. Woodbury, in February, 1874, the levels of 

 the Portland & Ogdensburg Railroad survey being found 5 feet higher 

 than those of our series along the Connecticut river. This disagreement 

 has been in part reconciled by adding one half the difference, viz., 2<| 

 feet, to our former figures beyond Dalton to Connecticut lake. 



Another series of levels, coincident with the Connecticut river line 

 from Groveton to North Stratford, is supplied from the surveys for the 

 Grand Trunk Railway. These heights were given as referred to tide at 

 Three Rivers, P. Q. To make them agree with our series when changed 

 as just mentioned, it is necessary to call tide at Three Rivers 30 feet 

 above mean sea level, which is quite near the truth, tide being stated 

 by Prof. Elias Loomis to be 9 feet at the mouth of the St. Lawrence, and 

 20 feet at Quebec, while it is probably considerably higher at Three 

 Rivers. Heights along this railroad have been accordingly referred to 

 mean sea level by connecting them with our first series. 



With the modification mentioned, Mr. Reed's levels gave the difference 

 between the Junction and Connecticut lake 1,249.626 feet, the height 

 of the lake being 1,618.863 feet above mean tide. 



