138 \V, .lOIIANNSKN 



of a synthetic hyl)ridizalioii Nvhen A and li or A -\- B and ('. were 

 brought together, the character »colour > or »hoariness» being thus 

 constructed». An analytical hybridization was realised when for in- 

 stance an organism with A -\- li -\- C was hybridized with, say, a -h b 

 + c. Here in the Fo-generation »analysis» of »colour and hoary» was 

 foud — as in Miss Saundehs' fine work. 



Results like these might have raised hopes as to a possibility of 

 segregating analytically the whole genotype into »factors) — and hence 

 in a remote future we- might be able to do some Homunculus-work 

 viz. to construct organisms through the addition or artificial combina- 

 tion of discreet factors, stored j)erhaps in l)ottles or small tubes!! 



But the nature of the genotypical units hitherto observed is highly 

 problematic. When we regard Mendelian »pairs», Aa, Bb and so on, 

 it is in most cases a normal reaction (character) that is the »allel» 

 to an abnormal. Yellow in ripe pease is normal, the green is an ex- 

 pression for imperfect ripeness as can easily be proven experimentally 

 e. g. by etherization. »No starch» in maize is evidently an abnormality 

 and so in the many cases upon which Bateson — as it seemed with 

 full reason — founded his for a time highly useful and suggestive but 

 now abandoned hypothesis of »Presence and absence»: the »normal» 

 almost always positive and dominant, the »abnormal» being (in a 

 morphological spirit) expressed as a »loss». 



Now the notions »normal» and »abnormal» in their valuing sense 

 are not adequate for Genetic analysis, hence classifications according 

 to such valuation are without interest. The question for us is this: 

 what is the nature of the difference between A and a. B and b and so 

 on? There is at present scarcely any doubt about the theory, that 

 »Mendelian factors» are in some way bound in or to the chromosomes. 

 The morphological view regards them as formed particles (say 

 »morphs», ad modum »allelomorphs») of the chromosomes, an old 

 Weismannian idea — mutatis mutandis. From a physiological stand- 

 point we may prefer to regard local conditions (say »chemisms») in or 

 on the chromosomes as responsible for those units — avoiding the 

 hairsplitting remark that »chemisms» are ultimately in some way » par- 

 ticulate ^ — as all things, even energy, now seem to be. 



If comparing an original (and in so far »normal») organism, for 

 instance a wild purple Lathyrus or a wild grey mouse, with the un- 

 doubtedly derived cultivated organisms, for instance a white sweet pea 

 and a yellow mouse, we might discover that there is one single geno- 

 typical point of difference between them, this difference may probably 



