10 BULLETIN 27, PORTO RICO EXPERIMENT STATION. 
phorus, and potash — gave only a small increase in combined average 
yield over the check plat in but one plat (that where nitrogen and 
phosphorus were given), and the plats where nitrogen or phosphorus 
was omitted gave actually a lower combined average yield than the 
check plat, it seems clear that no one element can be considered as 
the limiting factor in production, and that a complete mixture only 
should be used where fertilizers are applied. 
As the land utilized in this test had been in use for several years 
just prior to planting to the yams, and the yield was below the 
average usually obtained in Porto Rico, a better return was to be 
expected from fertilization than would be looked for from newer soil. 
In 1910 the four varieties used in the above-mentioned experiments 
were used in experiments to compare stable manure applied in differ- 
ent amounts with complete chemical fertilizers of varying amounts, 
and with a complete fertilizer made up of organic and chemical in- 
gredients. In these experiments stable manure was applied to two 
plats, one receiving the manure at the rate of 10 tons per acre, and 
the other at the rate of 20 tons per acre. Two plats were treated with 
a mixture of nitrate of soda, acid phosphate, and sulphate of potash, 
the mixture containing nitrogen, phosphorus, and potash in the pro- 
portions of 3, 6, and 9 per cent, respectively. One plat received three- 
fourths ton per acre of the fertilizer mixture and the other 1J tons. 
Another plat received 1 \ tons of a mixture having an analysis similar 
to that of the chemical mixture previously mentioned, but made from 
tankage, bone meal, and sulphate of potash. 
The quantity of roots harvested from these plats showed only small 
differences. No increase in yield was realized by replacing nitrate of 
soda and acid phosphate with tankage and bone meal. The plat to 
which 10 tons stable manure was given yielded nearly 10 per cent 
more than the check plat or the plats which received an all-chemical 
fertilizer. The plat which received 20 tons per acre yielded slightly 
less than the check plat. This heavy application resulted in a marked 
increase in the vine growth at an expense of the crop of roots har- 
vested. Ten tons of manure is sufficient for application. It may be 
in excess of the amount required for rich soils, since it is not spread 
broadcast as for most other crops, but is placed in ridges that are 
severel feet apart. When selecting lands preparatory to planting 
yams, and when using fertilizers on them, it must be borne in mind 
that the plants should make a thrifty growth if a good crop of roots 
is to be expected. Frail or extra heavy vines seldom give a good 
yield. 
TESTS OF VINE PRUNING. 
To test the effect of pruning on yams, plantings were made of 
Guinea and Potato varieties, because they are representative of large 
