The purpose of this paper is to present evidence of a recent de- crease in jacksnipe or Wilson's Snipe (Gcdlinago delicata), to the end of stimulating action for the conservation of this bird and its habitat. Its original abundance in the Mississippi Valley was probably beyond our present imaginative powers. Bogardus  (1874) killed 340 in a single day on the Salt Creek bottoms of the Sangamon River, and wagered to kill 100 straight in a day on this area. There were no takers. He says: "Our bag was seldom as small as seventy-five couple at the right time .... Snipe are vastly more abundant in the West . . . than in the East." Kumlien and Hollister 2 (1903) say of the jacksnipe: "still com- mon . . . [but] . . . we should be at a loss to express its numbers in former years." This refers especially to Walworth County, Wisconsin, where Kumlien began his observations about 1868. Schorger  (1929) gives the jacksnipe as an abundant migrant in Dane County, but states that "a gradual decrease in numbers has taken place during the last fifteen years." The extent of this recent decline may be roughly measured by means, of the following table and chart, compiled from Schorger's ornithological notes for 1919-1929, and my shooting journal for 1924- 1929. The table reduces the number of jacksnipe seen and killed by each of us to yearly averages of the number "seen per trip" (Graph A) and the number "killed per hunt" (Graph B). The reason for distinguish- ing "trips" and "hunts" is that Schorger made many trips during XField, Cover and Trap Shooting, A. H. Bogardus, J. B. Ford & Co., N.Y., 1874, p. 136. 2Birds of Wisconsin, L. Kumlien and N. Hollister, Bull. Wis. Natural Hist. Society, Nos. t-3, April-July, Milwaukee, 1903. aBirds of Dane County, Wis., A. W. Schorger, Trans. Wis. Acad. Science, Vol. XXIV, Nov., 1929. 184 The Wilson Bulletin--September, 1930 which no hunting was done. In both cases trips and hunts varied from a quarter day to a full day in length, and all were made in or near Dane County. My journal records the length of each hunt, so that I was able to reduce my figures to ternis of jacksnipe seen and killed per full day (Graphs C and D). Tile table, for simplicity, omits these calculations, but the greater smoothness of graphs C and D, as compared with A and B. reflects the removal of the disturbance due to varying lengths of time in the field. The downward trend of all four graphs is apparent at a glance. A median line, drawn by averaging coordinates in groups of three, has been added to Graph A in order to show its general trend, as distinguished from its temporary fluctuations. A downward trend is apparent in all the graphs except B, and is clear in this case when figures have been reduced to kill per full day on Graph D. The chart, therefore, indicates a progressive decrease in the abundance of jack- snipe in Dane County. Can this apparent local decrease be accepted as actual? If so, does it reflect a general decrease? A downward trend in birds killed might reflect poorer shooting rather than fewer birds. That there was no significant deterioration in my own shooting is indicated by data in my journal on shells per bird in bag up to 1926. As for Schorger's shooting, my impression is that it has improved rather than deteriorated. Both of us have used the same guns and I used tile same dog throughout the period covered. Even if there were no data on marksmanship, however, the downward trend of birds seen would still indicate a decrease. Moreover, the graphs make no allowance for increasing skill in where and how to seek birds. At the time our records begin I was new to the region, and Schorger had never hunted snipe systematically. That we have both learned something about their local habits is shown later on. In my judgment, even a horizontal trend in the various graphs would be reason for suspecting a decrease. Another explauation of the downward trend of all graphs might be that local shortages in food and water caused the migrating birds to pass over or around this locality. In so far as known, jacksnipe food is a function of water. The water in the remaining snipe marshes of Dane County is comparatively stable, because the marshes either lie at the level of artificially stabilized lakes, or are spring-fed. or both. Late summer and fall rainfall makes some difference, .even in spring-fed or lake-level marshes, but not nearly so much as in marshes fed entirely by river overflow or by rain. Rainfall figures for August Decline of Jacksnlpes in Wisconsin 185 ,)' 1/2K$/V/,øœ Gœœ/V  /.(//_LœZ I E C-GœEN PER DAy LEOPOLD  t(ILLFD $'EH I $OHORGER [] 0 , lEOPOLl) ß ß A- $œœN Pœfe TRIP, 1 AVœRAGE OF I '. CHORGœ,q g LEOPOLD 4o i...._ / , II LœOPOLO I '\ IO I i- ' -- L-HI;LœD PER HUNT! o  oo I I ? _ YEaR 13.7 . I. 43 12.07 I0. 8.7 I0. 10.91 .1 5. ]NCHEJ PRECIPITATION AGUST-OCTOBER INCL. FretroE 16. Graph showing diminution in nmnber of Jacksnipes seen or killed from year to year during these studies. 186 The Wilson Bulletin--September, 1930 to October of each year are entered at the bottom of the chart. I do not recall any year in which there was either a great shortage or a great surplus of snipe-water throughout an entire shooting season. Insufficient grazing might have reduced the attractiveness of our local snipe grounds, and thus account for an apparent decrease in birds seen and killed. It is not likely to have affected these figures, however, because Dane County is in the heart oF the Wisconsin dairy belt, in which both lhe number of cattle and the allocation oF areas used as pasture are quite constant from year to year. Snipe ground must ordinarily be grazed in order to be good, apparently because un- grazed ground does not offer enough exposed mud. The muddy or boggy shorelines of receding ponds are used when available, especially early in the season, regardless of whether grazed or no, but this ex- ception merely proves the rule.. Late November birds often resort to floating bogs covered with heavy ungrazed vegetation, but this is dur- ing cold weather. Apparently under such conditions the shelter-value of the vegetation offsets its obstruction of free access to the mud. Moreover, these floating bogs are then often the only ground left un- frozen. (These are all things Schorger and I have learned during the period covered by the graphs.. If the supply of birds had remained constant, this added knowledge should have produced a rising trend both in snipe seen and snipe killed). Drainage might bc another source of error. It is estimated that the available ground in Dane County has been shrinking at the rate of perhaps ten per cent pcr year by reason of new ditches. At this writing there are practically no large snipe grounds left except at lake-levels, where drainage can be effected only by pumping, and hence is seldom attempted. Numerous small upland potholes and spring- heads, however, still remain undrained. Ditched ground is usually worthless, even when wet by rains, and is avoided in hunting. Hence the only way for drainage to mave invalidated these figures is by switching the migration route. I cannot appraise the probability of such a change, except to say that since there is still enough ground to hunt on, there would appear to be still enough to detain a normal density of migrating birds. The actuality of the seeming decline in snipe is corroborated by the reports of local ornithologists on the spring migration, at which season the birds are not dependent on undrained marshes. For sev- eral years past the local bird-men, who each spring scour much country in search oF other birds while the hunters are not afield, have been reporting a scarcity oF snipe. Decline of Jacksnipes in Wisconsin 187 JACKSNIPE SEEN AND KILLED A. W. Schorger and Aldo Leopold, Dane County, Wisconsin, 1919-1929. Date 0-27 ............................................ 40 10-11 .......................................... 50 10-26 .......................................... 75 165 Seen per Trip ............................ 55 Killed per Hunt ........................ 1920 7-24 ............................................ 7 8-1 .............................................. 6 8-8 .............................................. 3 9-18 ............................................ 8 9-25 ............................................ 25 11-25 .......................................... 4 53 Seen per Trip ............................ 9 Killed per Hunt ........................ 1921 9-18 ............................................ 25 9-24 ............................................ 30 10-22 .......................................... 100 10-23 .......................................... 6 161 Seen per Trip ............................ 40 Killed per Hunt ........................ 1922 8-13 ............................................ 8 8-20 ............................................ 6 9-24 ............................................ 20 10-7 ............................................ 200 10-12 .......................................... 80 11-5 ............................................ 15 329 Seen per Trip ............................ 55 Killed .per Hunt ........................ 1919 Schorger Seen Killed x x 1923 Schorger Date Seen Killed 7-28 .................. 2 x 9-8 .................... 35 x 9-29 .................. 25 9 10-6 .................. 60 13 10-21 ................ 1 x 10-28 ................ 12 6 11-10 ................ 5 2 140 30 Seen per Trip.. 20 Killed per Hunt 8 L1/2opoh] Seen Killed x x x x x -- 1924 3 9-28 .................. 3 1 4 2 3 9-30 .................. 15 3 10-1 .................. 3 1 x 10-2 .................. 15 2 2 10-4 .................. 15 4 13 10-7 .................. 6 2 1 10-12 ................ 50 x 30 7 16 10-14 ................ 50 9 10-18 ................ 50 14 5 10-19 ................ 75 20 10-22 ................ 100 9 10-25 ................ 30 6 x x 10-26 ................ 20 6 x 10-28 ................ 60 9 12 11-4 .................. 40 11 13 11-9 .................. 7 6 x 158 27 417 85 25 Seen per Trip__ 39 30 12 Killed per Hunt 9 6 *The symbol x means no hunting done. 188 The V/ilson Bulletin--September, 1930 1925 8chorger Leopohl Date Seen Killed Seen Killed 9-13 .................. 1 x 9-26 .................. 4 x 30 11 9-27 .................. 20 4 10-10 ................ 1 1 40 10 10-11 ................ 9 0 50 14 10-17 ................ 50 11 10-18 ................ 25 2 10-22 ................ 50 12 10-24 ................ 20 2 25 8 10-25 ................ 2 0 10 1 10-31 ................ 8 1 50 9 11-1 .................. 20 5 25 10 11-7 .................. 12 7 77 16 315 92 Seen per Trip_. 8 29 Killed per Hunt 2 8 1926 8-28 .................. 5 x 8-29 .................. 3 x 9-19 .................. 14 x 9-21 .................. 6 1 9-25 .................. 7 1 20 7 9-26 .................. 9 x 12 3 10-2 .................. 15 4 10-3 .................. 3 0 10-4 .................. 3 1 10-7 .................. 20 3 10-8 .................. 20 6 10-9 .................. 35 5 10-10 ................ 2 0 10-15 ................ 7 4 10-16 ................ 2 1 1 1 10-17 ................ 30 5 30 6 10-23 ................ 100 11 10-30 ................ 30 6 118 22 256 43 Seen per Trip__ 11 24 Killed per Hunt 3 4 1927 Schorger Leopold Date Seen Killed Seen Killed 8-13 .................. 24 x 8-27 .................. 15 x 9-10 .................. 17 x 9-16 .................. ? 6 9-23 .................. 16 x 15 6 9-24 .................. 30 5 10-1 .................. 9 1 10-8 .................. 40 5 10-15 ................ 15 4 100 10 10-22 ................ 75 5 166 15 190 27 Seen per Trip.. 21 63 Killed per Hunt 4 7 1928 9-22 .................. 2 x 1 1 9-23 .................. 12 x 9-29 .................. 9 x 30 4 10-2 .................. 50 9 10-6 .................. 25 11 10-13 ................ 30 5 10-14 ................ 40 6 10-20 ................ 12 5 10-27 ................ 12 2 11-11 ................ 1 1 102 23 122 21 Seen per Trip__ 15 24 Killed per Hunt 6 4 1929 9-14 .................. 3 x 9-19 .................. 30 7 9-21 .................. 20 6 10-4 .................. 30 15 10-8 .................. 20 7 10-12 ................ 7 x 10-13 ................ 40 4 10-15 ................ 20 6 10-19 ................ 30 11 10-20 ................ 11 x 11-3 .................. 30 5 11-9 .................. 6 x 11-16 ................ 2 0 162 50 60 11 Seen per Trip__ 15 30 Killed per Hunt 7 5 Decline of Jacksnlpes in Wisconsin 189 All of the foregoing evidence pertains to Dane County and its immediate environs. In order to get a rough check on conditions else- where in the state, twelve sclected jacksnipe hunters, all from different counties, were asked for their opinion on recent trends. Of these, five reported no perceptible change in recent years, six reported a decline, and one reported that recent flights have been more sporadic than formerly. Taking everything together it is my conclusion: (1) that the jack- snipe in the region of Dane County, Wisconsin, has decreased perhaps fifty per cent since 1924; (2) that this may be due to their passing over or around us, or to a temporary abundance cycle, or to an actual decrease in the available supply; (3) that the only reason for doubt- ing an actual decrease in the available supply would be positive evi- dence that they have increased or held their own in the rest of the Mississippi Valley. If there is any such evidence of increase, I have not seen or heard of it. Such slight evidence as I have for the remainder of Wisconsin indicates that the decrease here indicated for Dane County has been statewide. The possible causes of the decrease are a matter of conjecture. One likely cause is the shrinkage in southern breeding ranges, which were possibly the most productive. Bogardus  says that jacksnipe formerly bred as far south as the Calumet River and the great Winne- bago swamp in Illinois, whereas Schorger a is in doubt whether they sill breed in Dane County. The twelve snipe hunters whom I ques- tioned concerning the status of jacksnipe elsewhere in Wisconsin re- ported their breeding in Sheboygan, Winnebago, Rusk, and Sawyer Counties. The most southerly of these is Sheboygan. From this, their present known southerly limit, to the Calumet River in Illinois, their probable southerly limit in 1874, is 120 miles. The only really comprehensive check against the further shrink- age of marshes would be to accord undrained marshes a special tax status in view of their public service to migratory birds, just as un- grazed farm woodlots and managed forests are beginning to be ac- corded a special tax status in view of their public value to watersheds and timber supply. Overshooting of jacksnipe doubtless occurs, but not so far in southern Wisconsin. The majority of hunters pay no attention to them as yet, but the number who do so is rapidly increasing. As nearly as I am aware, the diseases, parasites, and predatory enemies of the jacksnipe are unknown, and their food nearly so. An 190 The Wilson Bulletin--September, 1930 adequate life-history study would seem to be one of the obvious first moves toward a conservation program. NOTE: Since preparing this manuscript I obtained from Mr. D. H. Haines of Ann Arbor, Michigan, through the kind offices of his shooting companion, Prof. Kenneth McMurray of the University of Michigan, a digest of the former's Shooting Journal by days. This is summarized by years as follows: JACKSNIPE KILLED IN MICHIGAN BY DONALD H. HAINES Year Place No. Killed No. Hunts 1918 Kalamazoo ........................................... 68 21 1919 Kalamazoo ............................................ 47 26 1920 (Absent from State) 1921 Ann Arbor ............................................ 14 11 1922 Ann Arbor ............................................ 7 13 1923 Ann Arbor ............................................ 15 14 1924 Ann Arbor ............................................ 31 23 1925 Ann Arbor ............................................ 0 15 1926 Ann Arbor ............................................ 1 10 1927 Ann Arbor ............................................ 9 11 1928 Ann Arbor ............................................ 18 18 1929 Ann Arbor ............................................ 90 18 Total ................................................................. 300 180 Average ............................................................ 27 16 Mr. Haines' bag of snipe was obtained in conjunction with and sometimes incidental to a good deal of marsh duck hunting, hence his figures are not so direct an index to abundance as Schorger's or my own. For this reason they were not added to the graph. Nevertheless he assures me that whenever snipe were present in any numbers he usually hunted them. With respect to frequency, length, and regular- ity of hunts, his practice resembles Schorger's and mine. Mr. Haines' Ann Arbor bag was above average in 1924 and 1929. (The exceptionally high 1.929 figure was coincident with leasing some favorable marsh and hunting it oftener than usual). Our graphs show high in 1922 and 1924, and indicate an improvement in 1929. Mr. Haines' bag was low in 1922, 1925, 1926, and 1927. Our graphs show low in 1920, 1923, 1925, 1926, possible 1927, and 1928. The comparison is contradictory in only one year, 1922, and shows enough correspondence to suggest that Michigan and Wisconsin may both feel the same fluctuations in abundance. GAME SURVEY, MADISON, WIS.