32 



S ( " I K NT T F T r A G R I C IT L T t^ R E. 



January, 1921. 



tion to distribution of production so far as number of 

 months go. Crooked keeled birds show a somewhat 

 higher mean winter production than the straight keeled 

 birds or 38.9000 ± .7203 and 29.7220 d= .9366 eggs 

 respectively. The difference is not great but probably 

 significant. It has been suggested that crooked keeled 

 birds start to lay earlier as pullets than tliose with 

 straight keels. As has been stated before, over half the 

 birds used were finishing their first year's production 

 so the result obtained seems to indicate that this rela- 

 tion liolds true. 



Patterson and Quisenberry (7) state that straight 

 keeled birds laid more than crooked keeled birds or as 

 follows: straight keeled birds 100 per cent, slightly 

 crooked keeled birds 91 per cent, and decidedly crooked 

 keeled ])irds 68 per cent. These results are clearly con- 

 tradictory to those obtained in this experiment and 

 may be due to difference in stock or in management or 

 both. 



In order to get an expression for the curvature of 

 the keel three circles or arcs of circles were drawn 

 from one point, using as a radius the measurement 

 from the back 1o the antero ventral, midventral and 

 postero ventral surfaces of the keel. With a small rule 

 measuring to millimeters, a straight line was then 

 drawn equal to the length of the keel from the circum- 

 ference of the first circle to the last circle. From the 

 middle of this line, a perpendicular line was drawn to 

 tlie circle representing the distance from the back to 

 the midventral surface of the keel. This line was then 

 measured and used as an expression of the curvature 

 of the keel. The operation was repeated for each bird. 



There is a slight negative correlation of curvature of 

 keel with production as shown by the coefficients given 

 in Table 1. The coefficient of correlation in each case 

 is so small, liowever that in view of the probai)le errors 

 they cannot be considered significant. 



The get the relation of the length of keel relative to 

 the weight to egg production the length of keel was 

 divided by the weight and the result correlated with 

 production. This is shown in tables one and two. There 

 is a significant negative correlation here but it is worth 

 noting that this is caused largely by the position of a 

 few individuals. 



It is quite probable that these were abnormal as to 

 weight at the time thus changing the correlation suf- 

 ficiently to give a rather large negative coefficient. In 

 view of the result with the length of keel it seems some- 

 what doubtful whether this is of any significance as far 



Table 2 



as selection goes unless further work shows the same 

 result. 



In order to get an idea of whether the position of the 

 keel relative to the back was related to egg production 

 the measurement back to posterior end of keel was 

 divided by the measurement from the same point on 

 the back to the anterior end of the keel. The first 

 measurement is almost a direct measurement of depth 

 of body at that point but the latter is also affected or 

 modified l)y length of keel. This can therefore at best 

 be considered only an indication of the relative dis- 

 tance of the anterior and posterior ends of the keel from 

 the back. The expression thus obtained was correlated 

 with production and the result is shown bj' the co- 

 efficients given in table one. The coefficients of corre- 

 lation obtained are significant and mdicate tliat the 

 relative position of the keel may be related to produc- 

 tion but to establish this more direct methods and fur 

 ther data are required. 



From table 3 it will be seen that distance between the 

 posterior end of the keel and the pelvic bones is related 

 to total production. It ^^^ll be noted that tlie relation 

 is rather to length of laying period than to intensity of 

 production. Little correlation is shown i)etween the 

 size of abdomen and periods of production far removed 

 from the time at which the measurements were taken, 

 but there is a significant correlation to the last period 

 (July 1 — October 31) indicating that this measurement 

 varies with laying condition. To test this the birds were 

 divided into two groups, laying and nonlaj-ing, with 

 forty birds in the first group and two hundred and 

 fortv in the latter. The mean size of abdomen was 



5.7225 ±.0913 cm and 4.2788 ±.0386 cm respectively, 

 showing a decidedly significant difference which agrees 

 with results olitained by Kent (5). This variation 

 should be taken into consideration when selecting birds 

 by size of abdomen, though size of bird sliould also be 

 considered. 



Table 3. 



Coefficient of 

 Sul)ject Relative Correlation 



Size of Abdomen. . .Total egg production .2280±.0382 

 (Disli. Keel to 



Pelvis) Highest Montihly Egg Prod. 1439::p.0395 



No. of months laid 2346±.0380 



Egg Prod. Nov. 1— Fel). 28 .1338±.0396 

 Egg Prod. Mar. 1— June 30 .1398±.0395 

 Egg Prod. July 1— Oct. 31 .2095-I-.0385 



.00450— .00549 



.00550— .00649 



.00650— .00749 



.00750— .00849 



.00850— ,00949 



.00950— .01049 



.01050— .01149 



.01150— .01249 



.01250— .01349 



.01350— .01449 



2 11 28 44 61 76 38 14 6 280 



Length of Keel divided by "Weight of Bird, subject. Highest Monthly Egg Production, relative. 



Coefficient of Correlation == — .2896" ± .0369. 



