10 



only tell us with certainty which seeds would be able to germinate, 

 if the bird in question had been killed in such a way that the seeds 

 would come free: but it will also tell us with some probability 

 which seeds will ha ve a chance for dispersal in an endozoic man- 

 ner through birds. To get better knowledge concerning this låter 

 point, which is the one of importance, it will be necessary to 

 make other experiments. A. Keuxer (8) bas fed birds wilh seeds, 

 taken the unhurt seeds from the excrements and tested their 

 germinating power. Charles Darwin (3) has examined the excre- 

 ments of birds found in bis garden for seeds, and he reports to 

 have found 12 kinds of seeds, of which some germinated. This 

 låter way is the more natural one, but it has the fault that one 

 does not know the species of birds from which the excrement in 

 question came; further it is a somewhat occasional way which 

 will not give exhaustive results. 



The author has happened in finding once the excrements of 

 some water-bird, probably Anas hoschas, containing a lot of seeds 

 of Potamogeion naians. Next j^ear Mr. Ove Rostrup placed at his 

 disposal excrements of swans (Cygniis olor) stutTed with seeds of 

 Potamogeion notans just in the same manner. The author then 

 collected material of ripen Potamogeton-seeds at the same place 

 from where the swan-excrements came and began an experiment to 

 clear up the elTect of the passing through the digestive apparatus 

 of a swan. 



Both kinds of seeds were kept dry during a month, and then 

 four glass-vessels were filled wilh water to a height of about 10 

 cm.; in each vessel 100 seeds were dropped; in two vessels the 

 swan-seeds» the seeds taken from swan-excrements) were placed, 

 in Iwo others Ihc uiitouched seeds. The >swan-seeds» consist of 

 the bard endocarp and the embryo, while the untouched seeds fur- 

 ther have a lleshy exocarp outside. A couj)le of vessels consisting 

 of one with *swan-seeds» and one with untouched seeds was placed 

 during the winter close to the heating apparatus of a room; the 

 other couple slood between the double panes of a window and was 

 frozen at least from Decemb. 2(Slh to Jan. 5tli. 



The table on p. 6 gives the result of the experiment. It becomes 

 evident that the effect of the digestion has been in fa vour of 

 the germination. The »swan-seeds» germinate better and 

 faster than the untouched seeds. 



The author then discusses the manner in which this eilect has 



