66 FERN GROWING 



that had none of this previous cross. If the result were 

 obtained accidentally, Mr. Wollaston's idea might be correct ; 

 but if the same results are obtained over and over again, 

 this could not, by the same reasoning on the doctrine of 

 probability, be possible. Take the example of the mixing of 

 the four varieties of Scolopendrium (viz., a muricate, a crested, 

 a spiral, and an undulate form) : now, if only two of these 

 varieties effected the cross, and the other peculiarities were 

 due to previous crosses, how is it that the peculiarities are 

 restricted to these four varieties ? Why do we not get other 

 peculiarities that are spread through the hundreds of other 

 varieties of Scolopendriums, and how is it that, if we sow 

 from every one of the varieties carefully kept separate, our 

 seedlings are mostly a copy of the variety we have sown ? 



No matter how much different blood we have in a variety, 

 it is a fixed quantity, and may be treated the same as if it 

 were a single cross. 



A muricate and a normal but crested variety produce 

 muricate-crested forms, not all alike, but all muricate and 

 crested, and bearing a great resemblance to each other. 



If we add an undulate one amongst our spores, we have 

 a portion that are muricate, undulate, and crested, but by 

 far the greatest number are either muricate-crested, muricate- 

 undulate, or undulate-crested ; and if, in addition, we add spores 

 of a variety with a spirally twisted apex, we obtain a few 

 (very few) seedlings that are muricate, undulate, and crested, 

 with the crestings spirally twisted, a rather greater number 

 undulate, muricate, and crested, or undulate, crested, and 

 spirally twisted, but by far the greater number only a 

 combination of two of the varieties ; but we never get any 

 departure from the peculiarities of these four varieties,* showing 

 any other extreme forms that have not been sown. 



* I.e.. we only obtain some slight variation, and these are undoubtedly owing to 

 previous crosses. 



