57 



sig særdeles enkel, den fremgik o^jiindelig af faa celler, og 

 trods dette dannede den saa forskjelliij;e dannelser som 

 nervesystem og tarmtraktus. 



Yi fandt derfor i kap. 4, at knopanlægget lettest 

 kunde sammenlignes med primitive anlæg som f. eks. de 

 isolerede furingskugler i de udviklingsmekaniske eksperi- 

 menter. 



Med disse iiar nu kn(>i)anlæggene fremforalt det til- 

 fælles, at de 



1) frembringer et lielt individ bygget noiagtig som det. 

 der fremgaar af emLryonalndviklingen. 



2) at anlæggene forst synes „l3estemte" til en anden ud- 

 vikling. De celler, hvoraf den indre blære i knopan- 

 lægget dannes, er oprindelig et cellekomplex i et af 

 moderdyrets organer (f. ex. hos Botr//Uidernc i peri- 

 branchialsækken). Hvis man derfor ikke helt ud hyl- 

 der læren om „Die eingewickelten Keime-', saa maa 

 man antage, at dette cellecomplex ikke paa forhaand 

 er speciiiceret til dannelse af knopperne, men at det 

 faar stodet hetiil ander itdril.Tnigenx Joh. 



Knoi)udviklingen forekommer mig dei-for i ligc hoi 

 grad som de udviklingsmekaniske eksperimenter at vise, at 

 ikke al ndvikling fra første øieblik er specifikation. Og 

 knopudviklingen lærer os fremdeles, at en specifikation først 

 kan optræde meget sent i udvildinfien. Det er ikke alene 

 furliigsstadier, men ogsaa limhhidr og nUen-dc anlagte or- 

 ganer, hvorfra hiopanlwgget dannes, og disse kimUade og 

 organer maa derfor ialJefahl Uaudt sine celler kunne hevare 

 uspecificeret, indiffere)it cellemateriale. 



Paafaldende minder den af Canllerg (lUj beskrevne 

 dobbeltlarve, der tindes hos Diplosoma, om de eksperimen- 

 telt dannede doblieltlarver. Jeg afljilder denne lier efter 

 Caallei-g. 



Tile structare of tiie inner vesicle appearcd to be extremely 

 simple, originating from few cells; bnt in spite of that, it 

 fornied two such different formations as the nervous system 

 and the alimentary canal. 



We therefore found in Chapter IV tliat thi' rudi- 

 mentary bud could most easily be compared witii primitive 

 rudiments like the isolated segmentation spheres, for in- 

 stance. in developmental-mecluinical experiments. 



In eommon witli these, the bud-rudiments 



(1) produce an entire animal exactly similar in structure 

 to tiiat which proceeds from eml)ryonic development; 



(2) seem first destined for aiiother develojiment. The eells 

 from whicli tlie inner vesicle in tiie bud-rudiment 

 is fornied are originally a duster in one of the or- 

 gans of the pai-ent animal (in Botrgllidæ, for in- 

 stance, in the peribrancliial eavity). If, therefore, one 

 does not entirely iiold the doctrine of ,,die eingewi- 

 ckelten Keime''. one must presunie that this duster of 

 cells is not specified in advance for the formation of 

 tile buds, bnt that it receives the imindse to it in the 

 course of the derelopment . 



The devdopnient of the bud seeins to me therefore 

 to show just as plainly as the develoi)nu-ntal-mechanical 

 experiments, that all development is not specification from 

 the very beginning; and bud-development still teaches us 

 that a specification niay not appear until very late in the 

 det-elopment. It is not only the segmentatio)i stages, Ijut 

 also tiie germ-layers and tlie organs already comnienced, 

 from which the rudimentary bud is formed, and these germ- 

 layrrs tind organs mast therefore at ang rate he ahle to pre- 

 serve among their cells, toispecified. indifferent cell-matcrial. 



The double larva described by CauUery (10), whicii 

 is found in Diplosoma, strongly recalls the experimentally- 

 formed double larvæ. I here give an illustration of the 

 former after Caidlery. 



I-l. 



T = Thorax hos larven; x = thorax hos knoppen, <S' = larve- 

 hjernen ; H = halen med chorda, F = fæsteorganer. 



Salenshg har tidligere beskrevet, at denne doblielt- 

 larve dannedes ved deling; men CauUery har ved indgaaende 

 undersøgelser vist, at det ene individ, knoppen, dannedes 

 ud fra et primitivt anlæg ligesom hos de ovrige familier. 



T = thorax in the larva; t = thorax in the liud; S = larval 

 brain; ^=tail with notochord; J^= organ of attachment. 



Salenskg bad previously stated that this double larva 

 is fornied by division; Init Caidlery has shown by minute 

 investigations that tlie one animal, the hud, is formed from 

 a primitive rudiment, as in the other families. He found 



Den norske Nordhavsxepedition. .luli.in Hjort: Kimbl.idstiulier pa.a grundlag .if Aseidierues udvikling. 



