100 BULLETIN 1105, U. S. DEPARTMENT OF AGRICULTURE. 



Table 23. — Effect of brush cover on seedlhirj survival, ploiccd {/round, 

 Fort Valley area, 1914. 



Date. 



Number of seedlings per 

 square foot. 



Plot 1, 



no 

 brush. 



Plot 2, 

 brush- 

 covered. 



Plot 3, 



no 

 brush. 



Aug. 16, 1914 



Oct. 28, 1914 



June 30, 1915 



May, 19161 



3.1 

 2.3 

 1.9 



6.7 

 6.0 

 3.6 



8.5 

 7.4 

 5.1 



Summer, 1917 2 









Summer^ 1918 3. .. 









Apr. 19, 1919 



.14 



.70 



.54 



Survival from Aug. 16, 1914, to Apr. 19, 1919: Per cent. 



Brush 10.5 



No brush (average of 2 plots) 5.9 



1 No counts. Many seedlings broken at ground line by heavy snow of preceding winter. 

 " No counts. White grub killing all seedlings In patche.s. 



2 No counts. White grub active. Nearly all seedlings on east side dead or dying. 



This experiment demonstrates what may be accomplished by the 

 application of brush when associated with favorable conditions of 

 soil and moisture. It is fair to conclude that brush was in this case 

 beneficial to reproduction, but this does not mean that brush will 

 offset a deficiency in seed supply, a poor seed bed, and extreme 

 drought, such as were encountered in the 1908 experiment. It 

 should be noted that the brush was applied cafter germination. 

 Notes in connection with annual examinations show that the seed- 

 lings which were directly under the brush died, and the survivors 

 were in the small comparatively clear spots hetween the branches. 

 This relation is still apparent on the ground. (PL XVII, Fig. 2.) 



The experiment also indicates a beneficial effect due to plowing. 

 Studies in 1920, which will be described later, indicate that without 

 the plowing, which destroyed the native vegetation on this plot, a 

 lower survival would have been secured on the brush-covered than 

 on the open strips. On the seeded but uncultivated border around 

 the plowed plot very few seedlings were in evidence at any time. 

 Although this is not a fair comparison, owing to the inferior quality 

 of the seed on the unprepared ground, it is known that a good per- 

 centage of the old seed was viable, and therefore the results are at 

 least significant. In apparent contradiction to these results stand 

 those described under "Herbaceous vegetation," in which sample 

 plots 1b, 3a, and 3b showed less germination where the ground had 

 been cultivated than where it was left in the natural state. The 

 difference is readily explained by the fact that on sample plots 

 1b, 3a, and 3b practically no seed fell until four years after cultiva- 

 tion, with the result that the soil became packed, while in this 

 experiment the seed was sown and harrowed in immediately after 

 plowing. 



