833 



than ill its IV-llijws (if tlu' saiiii' ;ii;c. In ;i\(.'r;i;ii' enl- 

 culatiiJiis, liowever, sucii imlixidiial c'xcf|iti(iiis may he 

 regarded as eliininated, and iUv said classification is 

 observed, generally speaking, cxcn in ])oi)uIar ])hras('0- 

 logy, the first groiqi containing the La.rijiif/el, Salnion- 

 fry (lengtli under 7U mm.), the second groiij) the Sfi.r- 



70 mm. 



30 



rar, I'arr (length 



the third irroui) the Fo- 



rcller or Sfriihit (length ^ .,'.', j, tlie fourth group 



the OriiHiar or Borfii/ri lenii'tli ^ ~.l,. ' , the fiftli 



■ \ ' < 4o0 ,, / 



group tile full-grown Ld.rar or Salmon (length > 4oO 

 mm.). Tiie average percentages calculated as expres- 

 sions of the different relations of form in these groups, 

 lead us to very singular results". Out of 59 distinct 

 relations of form 41 follow a fixed course in their 

 changes of growth, i. e. they either decrease with in- 



er(^ising age (as for exami>le the a\erage length of the 

 iiead in propdrtion to tiiat of tiie l)ody), or increase 

 with age (as for example tiie average postabdominal 

 length in ])roiiortion to the length of the l)ody). 



In order to ol)tain an exjiression tor tliose common 

 variations of the Salmons whicli are due to external 

 sexual differences, we iia\'e cumpared 17 males of the 

 different stages of growth with 1 7 females of about the 

 same size as the males and with the sexual organs about 

 e(|uall\- ri])e. The result is that 111 of tiie said 11 changes 

 of growth run in the same direction as the sexual 

 difl'erence lietween male and female, i. e. a change of 

 growth that shows decreasing percentages with increas- 

 ing ;ige answers to a sexual difference witii greater 

 percentages in the males than in the females, and via; 

 rrrsd. The two examples cited above appear in the 

 following percentages: 



The jiarallelism between the numerical alterations 

 1) in the first five columns and '2) in tlie last two 

 columns, may evidently be expressed as follows: the 

 j'ounger Salmon are to the older as the males to the 

 females. 



If we now examine the difference of form between 

 tiie two Scandinavian Salmon species that have received 



most general recognition, Sahim tiiitta and iV. sahir we 

 arri\e at exactly the same result: the difference between 

 the averages for the said characters runs from the Gra- 

 la.r (S. fnitfa) to the Bhiriklax (S. salrir) in the same 

 direction as the a.\'erages for the changes of growth and 

 also for the external sexual differences. Some of the most 

 imp(n-tant of these characters may here be tal)ulated. 



" As we cannot give here all Ihe particulars of nur investigations, we refer tlie reader to our Kritish Fortechiimj 0/ver de >' Riks 

 mu$eum befinlliga Salmonidei; Vet.-Akad. Handl., Bd. 21, No. 8. 



