W. F. R. Weldon 
51 
JVote hy W. F. R. Weldon : 
The last two crosses had three olijects (1st) to see whether whites (produced by crossing- 
whites with first crosses) when crossed with first crosses— i.e. 1st generation hybrids — give the 
Mendelian numbers in their offspring, i.e. 50 % albinos and 50 % hybrids, (2nd) to see whether 
albinos produced in this way behave the same when crossed with waltzers as do ordinary pure- 
bred albinos, and (3rd) whether albinos whose ancestry has been "watered down" in this way 
for many generations will breed true when paired. With regard to the " watering down " of the 
albino ancestry of an albino, if hybrids at the first generation are paired with albinos, and 
the resulting albinos are paired back with hybrids of the first generation, tlie following tal)le 
shows what happens : 
The fractions in brackets after Ilyb. or Alb. indicate the amount of albino ancestry in the 
animal. 
Huh. (J) y^Alh. it =1* 
Alh. (f) ^Hyh. \ =\\ 
Alb. (I) xffyb. a =ii 
Alb. (^) xHyh. . II =1J^ 
Alb. (li) xffyb. =1.3^ 
Alb. (If) xffyb. Sf =1^; 
Alb. {^^)xByb. iM = lTk 
(BID X ^l'/^ 
From which it will be seen that to find the albino ancestry of an albino in a given generation 
the denominator of the fraction indicating the albino ancestry of the previous geneivation must 
be doubled to produce that of the desired generation : and to find the numerator of the gene- 
ration in question that of the previous one must be doubled and 1 siibti-acted. 
But it is evident, that continually breeding back with a first cross is bad: because the 
ancestry of the desired albino can never be reduced below a half. The first step (that taken 
in IJIQO to 91) is all right because the 5 lB9l is Alb. (|) and the yoiuig produced (i.e. 
iHdld}''^, etc.) will be alb. (|), i.e. an i less albino ancestry g—wf^'f ; then the degree in wliicli 
the " watering down '' diminishes is shown in the following series : 
XOO 144 ISn X-'S. 1J30 12!) 
2 5 0' 2 .■> 0 ' 2 .) (i ' 2 T) (i ) 2 'j (i ) 2 3 U 1 
or 16, 8, 4, 2, 1, — 
or ancestry of ? in 1//91, i.e. HQa- =-750 
„ „ 1//91«>, etc. =-625 
„ „ Albino in next generation = '562 
)> )) » 1) )) — '^'-'4 
» J> )) )) )) ~ '^l*^ 
)> )) )) >5 5) — 50/ 
7) n n » >) ~ 503 
that is to say starting with Alb. (|) (e.g. HGa'" in 1//91) we can in the 6th generation produce 
Alb. (-503). 
1//94 Parents doubtful, at any rate in colour of father and the 4 does 7/84« were pregnant 
when sexed. Offspring destroj-ed. — 1^95 The parent females IIQ2a etc. were pregnant wlien 
sexed. — l//95a''' Colour a mixture, between fawn, chocolate and c. — li7'95a-' Colour pale chocolate. 
— l//95a''' Killed by relations. — lZr96«" Colour a very pale fawn. — 1^97«' Pale fawn, witli little 
white on belly.— 1^98a* Much paler than a^—lHQSa^ Extremely like lH98a\ but with less 
white below. — iH'QSa''' A very blue fawn colour. — 1 H99a^ Bark f-Awn. The belly lighter but 
hardly any real white.— lZr99a2 Pale c. Skin not kept. — 1/?99«* Died young. — l/ZlOOa-* Skin 
not kept.— 1 fflOOfr' Rather dark fawn.— imoOa" Pale fawn.— 1 //lOlw' Colour a pale blue-grey, 
7—2 
