Discussion 
The ahove-fround transects represent the only data presently available for 
comparison with 1948 deta. It will be recalled that in 1948: a.} Several tran- 
Sect were run repeatedly and that the total representation of treeding pairs 
for any species was calculated for at least a portion of thoss transects, on tha 
basis of the maximum number of breeding paira represented on any ons coverags; 
b, soms transects were run sarlisr than in 1949 and as 9 tonsequencs this may 
have favored the 1948 dats slightly. The two aboys-noted dsyarturas from ths 
system preseribsd by the Fish and Wildlif* Services, although relatively insigni- 
ficant from tha standpoint of calcylation of the brssding density af waterfowl 
in Wiseonsin, may nevertheless represent an appreciable factor in messuring 
trends between 1948 and 1949. Therefore, although thse tetals on deta for the 
two years damonstrata an incrseass of only 0,03 breedine pairs per Square mils 
(or a 4 percent increase], actually tha increase may os markedly greater, 
Prasent projsct work is of such importanes that it precludes taking time for a 
Te-calculation of ths 1948 data. 
Bat on the basis of ths above considerationa and within ths gatatistical 
limits of the system used, and in view of th+ siza of ths sample taken, ths 
writer feels justified in stating that during 1949 thars has baen: 
(1) No changs in the 1949 Wisconsin waterfowl bresding population 
by comparison with data from 1944, 
(2) An indicated 4 psrcant increase in ths number of breeding palre per 
square mils. {Kot a valid figures on tke basis of the syatem 
usd, | 
(3) Good reason to believe that an upward trend much in excess of a 
4 pertent increase has taken placa. 
In #xémining the ahore data on specics composition only two conclusions 
arpeear warranted on thse basis of the data: 1.) That mallards have increased: 
2,) that blue-wirngesd tenl decreased. Present data in hend from ground and 
serial cenaus work conducted at Horicon Marsh also reflect the same trends in 
thease two species, tut the extent af these fluctuations has net been accurately 
measured as of this writing. An increases in slack ducks has also been indieatsd 
at Anrican. 
Watar Conditions 
Ground transects census werk offere the most rellable survey work on water 
conditions in the Stats, The late wintsr and spring of 1949 was characterized 
by a Limited smount of srnew and rainfall and relatively warm weather occurred 
throughout the neriod., Ground frost sapsrently disapnsarsd much earlisr than 
average and runoff water was limited in amount. Spring flood levels of farmer 
Foeers ware never reached on Hdoricon Marsh and as early as mid-April many tem- 
porary water areas had completely disappeared, Hy May 1 when consus work was 
begun temporary water arsas were almost non-existent. 
An analysis af field data on water conditions reveals ths followins: 
(1} There was a 28 percent decrease in the number of water argas. This 
decrease can be primarily attributed te the almost ecompliste dis- 
appearance? of temporary ponds throughout the revions csnsused. 
186 
