cri.M-;\ IM<.I:O\S ( UOSM.I) \\nii .<. \I\K.\ PIGEONS. S! 



owl-rock is the exception). l''urt hermore, those that did begin t.i develop stopped 

 sooner, producing birds of ;i shorter term of life ((! I homer of l!lli:; is a possible 

 though not ;i probable exception). Of the I!) eggs of Serie> I which showed anv 

 capacity for development, only one lived more than III day-, as compared with x 

 individuals from the 2( eggs of the other series which lived more than 7 months. 

 Thus it, is here clear that the series of "least fertility" is also the sen. 

 lived" offspring. 



In Series I, moreover, it is evident that when the female is more nearly pure 

 (('. tabellaria), not obviously hybrid uml rock), that both eggs of the clutch are 

 rarely fert ili/able; in none of the 12 clutches did both eggs show development . In 

 Series II, however, the 11 clutches (one clutch had but 1 egg) which showed devel- 

 opment showed it in both eggs of the clutch in 8 cases; 2 of the :; exceptions are 

 Jirst and last clutches of the season. The two series illustrate the fact that // /'* 

 when the developmental poirer of the germ* i* moxl xlronglu m///m/ tlml tin- 

 evidence becomes clearest for the developmental (</< <i/>/irl from xt.niol) 

 of the two eggs of the pigeon's clutch. 



When owl-rock hybrids are mated (4 mat ings) with the males of either series, 



the percentage of infertility is higher than when pure" InMlnrin or domedim 



is used, and at the same time the young that derelop from the hi/hrid union* have a 



.shorter term of life than have the young from female* of pure 



cfG 1. X owl-rock; 6 young lived 3, 7 days; 1, 5 days; 1. S days; 1, 111 days; averaue. 7 



<?G 2. X owl-rock; 4 young lived ;i,-7 days; 1, 10 day*; average, s day*. 



cfG 1. X homer; 8 you UK lived L', 1 day; 2, !l lo 10 days; X. M to li> days; 1, 15 mnntl .01 days. 



d*G 2. X homer; l young lived 1, 10 months; -. 1-+ months; :i, l.'() inonlhs; a\'-r;me, 170-H da\'--. 



cTG 2. X domestica; 9 young lived (i, 5 to 10 days; 1, 14 days; 1, 1 1 months: 1. :l |- mnnllm; avrraci', 142+ days. 



The data thus far considered lead, then, to the conclusions which are generally 

 supported by the data from other groups: (1) that fertility is a matter of all degi 

 (2) that proper crosses enable one to see and study the intcrgradations and flowing 

 levels of this important function; (3) that hybridization, even within the limits of 

 the breeds of common pigeons, makes their ova less compatible with the sperm of 

 one wild species; and (4) at the same time sets a sentence of a shorter term of life 

 upon the developments which proceed from their more compatible gametes. 



The non-equivalence of the two eggs of the clutch, particularly well seen in 

 cases where there can be but little development (as in (/ / /' .' . has also been 

 indicated. It remains to point out that these data further show that germs pro- 

 duced at the extremes of the seasons are weak in developmental power: fewer of 

 such eggs develop, and shorter lived offspring arise from them. This result can 

 here be best presented 7 in a table showing the outcome for the eggs of each month 

 of the year. Eliminating the record in Series II for the whole years in which no 

 fertile eggs were produced, the data are as shown in table (i(i. 



It will be seen that the "highest fertility " and "longest life" are both resident in 

 the eggs produced in May. The second for fertility is April; the second for longevity is 

 March. The lowest fertility is found in December, January. I'Ybruary, and August. 



'The (*KX (if urc'atcr dovclopincntal slrciinlli ancl of nuk-producing tcudi-ury srcins in ilninr.xlu- piiliiilis. i-vrn 

 of so-called "pure strains," to bear no definite relation to the order of the eji.ns in the clutch. This is as in hybrids 

 generally. 



6 By "pure" of course only "relatively pure" is meant when speaking of common pigeons. 



7 All such methods of grouping and treating data not entirely homogeneous are of course open to objection 

 but an estimate of the situation under analysis can be thus presented. 



