TERATOLOGY IN MODERN BOTANY. 95 



mally arranged in four rows, the leaves in two of the rows 

 being- much smaller than those in the other two rows. An 

 arrangement of leaves such as is exhibited in these gall- 

 bearing shoots (which are inhabited by the larva of a 

 Dipterous insect) is never found in Selaginella under other 

 circumstances. The bulbils grow with a triangular pyra- 

 midal apical cell. The first stages in their development are 

 unfortunately unknown, but they probably begin as out- 

 growths on young shoots as a result of the irritation pro- 

 duced by the larva, 



A similar instance is afforded by the singular "witches 

 brooms," which are formed upon the leaves of Pteris 

 qiLadriaurita^ as the result of the action of a mould, 

 Taphriua coi^muervL These are adventitious shoots with 

 misshapen leaves ; adventitious shoots do not normally 

 appear upon the leaves of the plant in question. 



The malformations, resulting from the action of animals 

 already alluded to, lead us to glance at the latest observa- 

 tions upon the origin of galls. It is not surprising that 

 these curious structures should have attracted attention at 

 an early period. We see here that, as the result of irritation 

 applied by an animal, there arises a new growth which is 

 admirably adapted to afford food, dwelling, and shelter to 

 the creature therein developed. What is the nature of this 

 stimulus? In the opinion of Lacaze-Duthiers,^ with which 

 Darwin and Hofmeister also concur, a poisonous substance 

 is introduced by the insect when laying the ^'g'g by means 

 of its ovipositor, and that this gives rise to the formation of 

 the gall. It is evident that the irritation to which the parent 

 insect gives rise presupposes a power of responsive action ; 

 the virus of Cynips j'oscr, for example, is destitute of action 

 upon the oak, which nevertheless so readily produces galls 

 under appropriate stimulation. 



As has, however, been shown by the well-known 

 entomological researches of Adler, the theory enunciated by 

 Lacaze-Duthiers cannot be regarded as trustworthy. In 

 the department of botany our knowledge of the formation 



1 " Recherches pour servir a I'histoire'des galles," ^;/;/. Sci. Nat. Bot., 

 1853- 



