294 Beer, Development of the iiollen grain and antlier of some Onagraceae. 



red leaves it unstained. JMulliyli'iR-liluL' and i'uelisin mixture 

 stains thu layer })ink or violet. 



I have found tliat the first pollon-Avall of the Horse- 

 chestnut in its early stages, <j^ivos reactions which are ahnost 

 identical witli those of the secondarv hiycr of Ocnotliera. Apart 

 from the very striking violet reaetion with the lodine sohition 

 the properties of these membranes correspond fairly well with 

 those characteristic of pectic substances and it is not improbable 

 that we are here dealing with an association between a pectic 

 body and a substance oi' unknown nature. Additional support 

 is given to this view by the fact that the violet reaetion be- 

 comes lost after treatment of the pollen grains with absolute 

 alcohol, no donbt because the body which gives this reaetion 

 is soluble in alcohol. In its behavionr towards other rcagonts, 

 however, the thickening layer remains unaltered after an immer- 

 sion in alcohol. Cuticularisation takes place very early in these 

 membranes ^j and the violet-reacting body may he associated 

 with the first stages of this process. 



In alcohol material the thickening layer, at the early peri- 

 ods of its development, is often greatly swollen -) and this be- 

 comes more markcd and may even lead tp the bursting of the 

 pollen grain if this be examined in aqueous Solutions. 



The interstitial body is now^) limited towards the cavity of 

 the grain by a closing disc which has the same composition as 

 the rest of that body although it is somewhat denser. The 

 reactions of the whole interstitial body have undergone a change 

 and are no longer those of a pure pectic body. With lodine- 

 reagents it colours yellow: with congo-red it stains uniformly 

 red; with naphtol black it colours blue-black; with nigrosin it 

 becomes black ; with ruthenium red it stains red ; with methylene 

 blue-fuchsin mixture it colours blue, pink or violet according to 

 the strength of the Solution used; with coraUin-soda Solution it 

 remains colourless. 



The protoplast fills the cavity of the pollen grain at this 

 stage but weak plasmolysing agents show that, whilst it is firmly 

 fixed to the developing secondary layer, it is free from the bases 

 of the interstitial bodies. 



As the thickening layer of the pollen wall continues its 

 develoj)ment ring-shaped ridges make their appearance at the 

 bases of the interstitial bodies. These are at first low and in- 

 conspicuous but soon become sharp and prominent features on 

 the membrane (Figs. 23 and 24). 



In pollen grains of Oenothera lojigiflora '^j which measure 



1) After whicli they colour yellow to brown with lodine in potassiiim 

 iodide Solution. 



-) This was already noticed by Strasbiirger in Gaura biennis. 



3) Pollen grains measure at this time 35 to 38 ^ in Oenothera biennis 

 and 0. longiflora. 



^) Although I give here the actual description and measm-ement of 

 the pollen of Oenotliera longiflora the facts are essentially the same in 

 0. biennis. 



