74 rKocKii:i)iN(;,s or rui; 



:i(lii)ir;ibl(' allcniative, wliicli may even confer a new lease of life 

 on tlK> ^took wliuh employs it. 



ir tilt' ■ T/iw of iioss' be aL'ee|)ted as ot" general application, it 

 fiM'iiislii's :i cliii' to certain phylogenelic problems. It throws 

 some lifjlit, For instance, on such diflicult questions as the inter- 

 pretation of tlie flower of Hidas. If we take the usual view that 

 y^aias ii a highly evolved and reduced member of the Potamo- 

 Retonacea? sto-k, we have to suppose that tlie immediate ancestor 

 of the gpnus possesseil no perianth, since a true perianth seems to 

 be unknown in this family. According to. the 'Law of Loss,' it 

 MOiilil he impossible under these circumstances for this genus to 

 evolve a true ])erianth, and hence we are relieved from the 

 necessity of trying to interpret any of the floral envelopes of 

 JV«m.«- which show very little indication of foliar characters — 

 as if they were of perianth nature. We are thus free to suppose 

 that the delicate, membranous floral envelope which Xaias has 

 evolved, is merely a cii]>uhir outgrowth fi'om the receptacle, while 

 the ])etal-like wings of tlie stamens in Potaiuof/efon mav represent 

 anotlier method of replacing the pi-rianth wliirh, having been 

 losr, cannot be resuscitated. Again, it is highly nnlikely, on the 

 * Law of Loss,' that a naked unisexual flower could evolve into a 

 h.-rmaphrotlite flower with a perianth ; hence the Law accords 

 wirn the view that the ]n-imitive Angiospermic flower was of such 

 a type as that met with among the Ranales, rather than with tlu^ 

 opposed theory that it reseml)]ed one of the floral forms found 

 among those Archichlamydeaj which were formerly called Apetalae. 



In the case of the Angiosperms wt- possess at present no fossil 

 evidence throwing light on the pliyletic lines within the group, 

 a,nd all our theories on this subject nuist, for this reason, be based 

 upon comparative morphology alone. Tlie zoologist, on tlie otlier 

 hand, is more fortunate, since abundant material has been unearthed 

 in recent vears, bearing on the paUeontological history of the Verte- 

 brates — a group which occupies a ]>lace among animals comparable 

 in some respects with that of the Angiospeims among ]ilaiits. 

 Some time after I liadderluced the Law of Loss from a considera- 

 tion of the structure of the water-plants living to-day, I learned 

 that zoologists had alreadj' arrived, on fossil evidence, at very 

 similar conclusions concerning Vertebrates. The brilliant iielgian 

 palfeontologist, Dollo*, enunciated in 1893, in tliH form of a 

 series of aphorisms, certain laws of evolution which he tells us he 

 reached as a result of studying fossil Vertebra'es for a number of 

 years in the iJrussels ]\[useum. He states that " L'evolution est. 

 discontinue — irreversible — limitee." lie explains what he means 

 by irreversibility by asking — and answering in the negative — the 

 question " Un Organisme peut-il retoiirner (totalementou partiel- 

 lement) ii uu etat anterieur, deja realise dans la serie de ses 

 ancetres?" DoUo's dictum regarding the irreversibility of evo- 

 lution was expanded and continued b}' various workers and in 

 particular by Osboruf in his Ixiok on ' The Age of Maininals.' He 



* Dollo, L. (1893). 



t O.iboru, H. F. (1910) ; see also Woodwarcl, A. Smitli (1808), etc. 



