326 LINN^US AND THE JUSSIEUS 



to. Some old-established natural groups were thus 

 preserved. Linnaeus avoided the association of the 

 Kosacese with the Ranunculacese, and kept the Labiates, 

 Crucifers, Composites and Orchids (with some few ex- 

 ceptions) free from mixture with alien genera. Not 

 even his own class-definitions could induce him to 

 break up the Leguminosse into Monadelphia and Dia- 

 delphia, though he separated the decandrous genera. 



The real test of any classification of living things 

 is : — What sort of groups does it yield ? Some of the 

 Linnean classes and orders were soon seen to be un- 

 natural. To separate Anthoxanthum, Holcus and Zea 

 from the Grasses, Veronica from the Scrophularineae, 

 Salvia and Lycopus from the Labiates, Sanguisorba, 

 Poterium and Alchemilla from the Eosacese ; to associate 

 Mercurialis with Hydrocharis, Valerian with the Irids, 

 Potamogeton with Holly, and Arrowhead with Oak and 

 Beech ; to make an order out of Paris, Adoxa, Elatine 

 and Sagina, and a class (Polygamia) out of Musa, Vera- 

 trum, Holcus, Atriplex, Chamserops, Morus, Fraxinus, 

 Rhodiola, Empetrum, &c. must have troubled Linnaeus 

 himself. Four of his twenty-three classes are natural 

 assemblages, but all these had been recognised before 

 under other names ; two more might have been rectified 

 so as to become tolerable ; the rest were unnatural. 

 Linnaeus could find consolation only in the facility of 

 his system, which was almost indispensable to the 

 progress of systematic botany at that time. How a 

 handful of European botanists could have dealt with 

 the continual accession of new species from every 

 quarter of the globe, if they had been forced to ponder 

 difficult questions of affinity at every turn, it is not 

 easy to conceive. 



Facility of course became less indispensable in time, 





