2. PROCEDURE 
In order to expedite the search for data sources, the study was di- 
vided into two basic parts: the hydrographic and the meteorological. 
This was a natural division since the bulk of the meteorological data 
was expected to originate at points apart from the sources of hydro- 
graphic data. However, it was known that certain agencies obtaining 
routine hydrographic data also obtained concomitant meteorological ob- 
servations. In such cases, it became the responsibility of the per- 
sonnel in the hydrographic division of the study to ascertain the neces- 
sary information relative to the meteorological observations, and to 
then transmit it to personnel in the meteorological division. The pri- 
mary reason that many meteorological sources are different from hydro- 
graphic sources is because it was deemed necessary to obtain meteoro- 
logical data not only around the periphery of the Lakes, but inland for 
some distance as well. The influence of the Lakes on weather conditions, 
and the influence of weather on the Lakes, is known to encompass an area 
around the Lakes as well as over the Lakes themselves. The exact limits 
of this "area of influence" are yet not completely determined, but for 
the purposes of this study have been confined to the drainage area of 
the Great Lakes (Fig. 7). 
The first effort by project personnel to locate all pertinent 
sources of meteorological data within the Great Lakes basin was made by 
contacting the National Weather Records Center of the U. S. Weather Bur- 
eau at Asheville, North Carolina, and the Meteorological Division of the 
Canadian Department of Transport in Toronto, Ontario. These two agen- 
cies provided project personnel with information on meteorological data 
that is published. This comprised the largest source of all types of 
data uncovered by the project: 808 sources or 68.6 per cent of the 
total of 1177 sources (see Table 4, p. 133). 
All other meteorological data sources ascertained by the project are 
comprised of unpublished, unprocessed data on file at each station or a 
central repository. The data are recorded by U. S. Coast Guard Stations 
(some of the data from a few of these are published in U. S. Weather 
Bureau climatological summaries), water treatment plants, industries and 
power plants, sewage treatment plants, paper mills, commercial and re- 
search lake vessels, and a few other sources such as university research 
groups, individual observers, and governmental and public service organ- 
izations. 
The search for hydrographic sources was initiated by concentrating 
first upon the water treatment plants. Information concerning data a- 
vailable from such plants in the United States was obtained by contacting 
the head offices of the public health departments of the states bordering 
the Great Lakes: Illinois, Indiana, Michigan, Minnesota, Ohio, Pennsyl- 
vania, New York, and Wisconsin. In Michigan and Ohio, at least a portion 
of the data from these plants was found to be available from the head 
offices, where it is kept on file. In the other states, data are re- 
tained in the files of the individual plants, from which they may be ob- 
tained. Information on water treatment plants in Ontario was furnished 
by the Ontario Water Resources Commission. 
Another source investigated early in the study comprised the power 
plants which utilize water from the Lakes. A list of all such plants on 
