124 REVERSION AND ALLIED PHENOMENA 



the inheritance and are really transmitted from generation to 

 generation, may be kept under by other components of the 

 inheritance, or in some way prevented from asserting them- 

 selves. At length, in the reconstitution which is associated with 

 the maturation and fertilisation of the germ-cells, or in the 

 intimate germinal struggle which is possibly always going on 

 amongst the diverse hereditary items, the long-latent items find 

 their opportunity and the result is a reversion due to the re- 

 assertion of long-latent characters. The progeny of the fancy 

 pigeons spoilt by hybridising harks back to its distant ancestor, 

 the rock-dove. 



Thus conceived, reversionary phenomena are simply illustrations 

 on a big and striking scale of what is familiarly observed in the 

 common phenomena of " taking after a grandfather " or " skipping 

 a generation." The difference between the case of a boy who 

 has the eyebrows of his great-grandfather, and the case of a 

 domestic pigeon which harks back to the ancestral rock-dove, 

 is simply one of degree. The difference between the occurrence 

 of white spots in some lambs of a herd carefully selected for 

 black wool, and the occurrence of stripes on the forequarters 

 of a Devonshire pony, is simply one of degree. All illustrate 

 ancestral inheritance. But the difference in degree is so marked 

 that a special term — " reversion " — is convenient. 



The garden of a shepherd's cottage swallowed up in a deer- 

 forest lost all trace of its previous cultivation and became a 

 weed-ground. After many years it was delved, and soon there 

 appeared many different kinds of old-fashioned flowers whose 

 seeds had lain dormant for several generations. So may ancient 

 flowers and weeds now and again reappear out of latency in 

 that garden which we call our inheritance. 



§ 4. Phenomena sometimes confused with Reversion 



It is impossible to read the fairly abundant literature without 

 becoming convinced that many phenomena are labelled "rever- 



