— 22 — 



In insula St. Jan Indiae occidentalis. (Leg. Raunkiaer). 



The fungus above described was found in a brittled trunk, occupied 

 by ants, and on the label is written: "Ant-nest with fungus-culture on 

 the walls of the walks". In reality there is good cause to presume that 

 this fungus is cultivated by the ants. According to Raunkisr's obser- 

 vation all the ant-walks in the wood were covered all over with the 

 brown-yellow layer of conidia and outside the walks the fungus was not 

 to be seen. This does not prove, however, that the fungus is cultivated 

 which only becomes evident when we are comparing the case with Al- 

 fred Moller's investigations on the relation of certain ant-species to 

 fungi ^). His examinations prove that some ants (of the genus Atta a. o.) 

 know how to cultivate certain species of fungi and how to carry on the 

 cultivation in quite a definite direction. Besides this they know how 

 to weed the culture of all weeds consisting of foreign mycelium, bacteria 

 and the like, consequently they are making these "fungus-gardens" models 

 -of pure-cultures. Furthermore Moller's experiments have proved that 

 fungus-cultivating ants are eating nothing but their own fungus and have 

 such a love of this food, that only on the point of starving they will 

 try some other food. 



Comparing our here mentioned finding with these experiences it 

 seems at hand to suppose that the ants, when they are putting up with 

 the fungus, are finding their account in it. — We are to suppose that 

 the conidia of the fungus that are so to speak stuffed with oil are to 

 the ants a good article of food. (Compare the oilrich "Kohlrabihaufchen" 

 in the South-American ant-species' fungus-gardens). 



St. Jan: America Hill 19.3.06 (nr. 1782). 



Chromosporium pachyderma Ferd. et Wge. sp. n. PI. II, fig. 12. 



Conidiis globosis, subglobosis angulatisve, episporio usque 4 /i in- 

 crassato, 15 — 20/^ diam. , viridi-pellucidis, supra lignum decorticatum 

 crustam tenuem, aerugineam, late effusam formantibus. Mycelio par- 

 cissimo, conidiophoris nullis. 



Ad hgnum decorticatum in insula St. Croix Indiae occidentalis. 



St. Croix: Cresquis 4.1.06 (nr. 1639); Jolly Hill 16.1.06 (nr. 1647). 



Fumago vagans Pers. Myc. Eur. I, p. 9. 



This fungus appears, at it seems, often in The Danish West Indies 

 on many different plants; most richly developed we found it on Mangifera 

 indica, but just here it was also noted that the trees in question were 

 "infested with plant-lice in the top". We have noted all the known 

 forms of fructification of the fungus with the exception of simple conidi- 



^) Alfred M0ller: Die Pilzgarten einiger siidamerikanischen Ameisen in: 

 Schimper: Mittheilungen aus den Tropen. 



