14 
BULLETIN 3 4, PORTO EICO EXPERIMENT STATION 
However, between the third harvest in July, 1913, and the forty- 
fifth harvest in January, 1924, the record was broken by the omission 
of only three harvests — the twenty-second, the thirty-eighth, and 
the forty-third. 
Figure 6 shows the average production per palm per harvest. 
The solid line shows the average production if all the palms are in- 
cluded, whereas the broken line shows the production averaged at 
each harvest for only such palms as contributed to that particular 
harvest. The curves very closely approximated one another for 
the first seven years, after which they diverged widely. This dis- 
similarity in the last four years indicated one or both of two things : 
Either many palms produced nothing at all or the pickers were 
unwilling to climb palms which offered only a small return for the 
labor involved. In either case a diminishing production was shown. 
The pickers, who receive so much per thousand nuts thrown down, ' 
will not climb a tall palm which offers in return for their efforts 
only one or two mature nuts. The curves show that whereas the 
Nuts 
li 
it 
13 
IZ 
II 
10 
9 
6 
7 
6 
S 
♦ 
3 
1 1 1 1 1 1 1 1 1 
Indiriduol harvests 
1 i 
I 
1 
1 
ri i i ■ 
Harrtsts by periods 
1 
— _ 
m Average number of nuts per po/m. inclusive of nonproducers 
j 
| 
| 
1 
1 
1 
»- 
) 
\ 
* 
\ 
i" 
\\ 
\ i 
^ 
*\ 
\ 
II 
if 
V 
t 
* 
i 
i 
-i 
:\ 
/ 
t 
i 
1 
1 
: 
• 
1 
\ _ 
\ 
; 
* 
_ i. 
*i 
if 
f 
1 
/ 
i**--. 
// 
1 
If 
if 
\, 
i 
1 
V 
•-' 
\ 
i 
I 
i 
1 
Y/ 
\y 
Vj/ 
X'l 
I 
1 
\ 
''-If- 
i 
i 
\ 
. 
/ 
/ 
\ 
\ 
\ 
\/ 
\ 
1 
\ 
\ 
■*-«. 
\ 
1 
/ 
finnf4^<nmviuticifi<o<OKKM 
«1 « 'N*nfl)^MtOB^"*)*OOt^') 
»K KO| •>-- -.♦ 
•< 0«O p»o KOO nM 
period Mo 
/*J*S6769I0II 12. 13 lolS 16 17 If ISZOZI ZZZ3ZH iSlill at33Q3l 31333H3S3LS7 3»39<lCil *l+S**itS 
J 
s 
Figure 6. — Average yield of nuts per palm in harvests considered both individually 
and grouped, Boquillas plantation 
planter suffered a heavy loss, the pickers did not allow this diminish- 
ing production to affect their pro rata receipts from palms climbed, 
which remained approximately as formerly. Had the pickers been 
willing to bear a share in the reduction of returns by climbing all 
the palms bearing mature nuts, the reduction in yield would have 
been less pronounced, the income to the planter would have been 
greater, and the record of production as influenced by soil treatment 
more satisfactory from an experimental point of view. 
From the data presented in Figure 6, a seasonal production curve 
was constructed as is shown in Figure 7, each point being determined 
by averaging the collections of the preceding and of the following 
month with those made during each month. This production curve 
showed that the peak was reached in January and that the smallest 
production was made in the summer months. 
The comparisons of individual and plat yields were based on the 
40 harvests recorded between July, 1913, and January, 1924. For pur- 
poses of analysis, the harvests were divided into five periods of eight 
harvests each. The first, second, and fourth periods were unbroken 
