Some Observations on the Genus Kubus. 177 



5. When new forms are created, there appear in them not only such cha- 

 racters as are »iure or less directly called forth by the influence of outward 

 circumstances, bui also characters which have their origin in dispositions latent 

 in the parent species. These dispositions can be partly inherited (page 45), 

 partly acquired, although they hare not yet appeared in the externat organisa- 

 tion oj the parent species (pages 9, 104). 



Several examples previously have been shown of the fact, that inherited 

 dispositions which have been suppressed in a form, can again show themselves 

 in the otiter organisation of a new form developed from it. And ic is not an 

 tincommon occurrence that the disposition which has not yet been shown outwardly, 

 in a species, can produce outward changes in the forms, that are developed 

 from it. Thus, in R. corylifolius *maritimus characters are found again 

 which are wanting in the species from which it sprung, but which are again 

 found in Ii. suberectus (page 104). In R. bahusiensis, var. nitens charac- 

 ters ariae which also distinguish R. sulcatus, and R. * bahusiensis, var. 

 obuinbratus has characters corresponding to R. fruticosus. H. suberectus 

 has a variety conjungens which reminds one of R. sulcatus (page 1C> 7) which 

 is developed from 11. fruticosus. And if my conjecture is right, that both R. 

 idaeus and R. suberectus were indirectly developed from R. saxatilis or R. 

 caesius, then R. ida'iis also can serve as an illustration of this rule. For as 

 has already been shown, the terminal leaflets on the turiones of the R. idseus 

 BOtnetimes resemble, in a great degree, those of R. suberectus. And when K. 

 Lindebergii which originated from R. insularis. produces the variety sericeus 

 (page 14.">) which answers to R. polyanthemos, produced by R. insularis, so this 

 must be ascribed to dispositions, that" had not yet appeared in the external 

 türm of R. insularis, when R. Lindebergii arose out of that species, and 

 only afterwards showed themselves in R. polyanthemos. 



6. From the facts given in the tffth thesis concerning the origin of nerv 

 forms we see, that analogy is a kind of likeneßs, that flors not slum! in any 

 Opposition to affinity, 01 the likeness that depends on common origin. Indeed, 

 the more nearly related Iwo series of forms are, still greater is o/so the ana- 

 logy between the forms in both series. And in the same degree, the more di- 

 stant the relationship is, so the likeness, founded on analogy, is less and at los/ 

 almost accidental and confined to Single characters. 



I have already given manv proofs as to the correctness of this statement. 

 Thus, for instance, both R. *nemoralis and * Wahlbergii, two subspecies 



