HORTICULTURE. 147 



application to garden peas. All other seeds tested were sterilized by either 

 period of application of hydrogen peroxid except in case of a 1 per cent solution 

 on radishes. Formaldehyde or 0.5 per cent mercuric chlorid almost invariaijly 

 sterilized, but 0.1 iwr cent mercuric chlorid completely failed in case of peas and 

 beans and in the shorter or 5-minute application to alfalfa and corn. Treat- 

 ment with formaldehyde and mercuric chlorid resulted in distorted sprouts when 

 the seeds were germinated but affected hard-coated leguminous seeds less than 

 the nonleguminous seeds, most of which were killed. 



The residual effects were tested by attempts to grow Bacillus suhtilis in wash- 

 ings from the seeds after 2, 3, or 4 rinsings. Two rinsings failed to remove 

 hydrogen peroxid but 3 were sufficient. Seeds treated in vacuum were more 

 perfectly penetrated by the sterilizing agent, but this advantage was more than 

 counterbalanced by the greater difficulty experienced in removing its toxic 

 effects before inoculation. Seeds dried in soil had a much lower number of 

 bacteria 1, 2, or 4 days after sterilization. Culture-treated seed previously 

 washed with sterile water were lower in number of living bacteria 1, 2, or 5 

 days after treatment when dried in soil or filter paper than in desiccators or 

 open Petri dishes. The author concludes that careful seed disinfection should 

 precede legume inoculation experiments and that metallic poisons such as mer- 

 curic chlorid are objectionable because difficult or impossible to remove from the 

 seeds. Sulphuric acid is more readily rinsed off than mercufic chlorid, while 

 hydrogen peroxid is still less harmful in its residual effect and is effective in 

 eliminating bacterial contamination. 



Seed testing apparatus. — A study of conditions under which our germi- 

 nation tests are made, H. Garman (Kentucky Sta. Bui. IJfS, pi). 41-62, pi. 1, 

 figs. 8). — This bulletin contains two papers. 



I. Seed Testing Apparatus. — For blue grass and other small seeds, excellent 

 results are being obtained in germination tests by the use of an ordinary jelly 

 tumbler on top of which is placed a bobeche or candle-drip glass. The seeds 

 are placed in this glass on a disk of blotting paper which is kept moist by a 

 candle wick leading to water in the jelly glass. For the purpose of controlling 

 air pressure in removing foreign seeds and inert matter from pure seeds, a 

 mercury gage registering 2 lbs. pressure was found helpful. It facilitates the 

 determination and maintenance of the pressure needed to remove impurities. 



II. A Studi/ of Conditions Under Which our Germination Tests are Made. — 

 Results reported with grass seed samples kept in the dark in the headhouse in- 

 variably showed a lower germination test than those treated in any other way, 

 but with clover and alfalfa seeds the differences were of no consequence. 



The control of quack grass and Canada thistles, A. L. Stone ( Wisconsin 

 Sta. Circ. Inform. 19, pp. 13, figs. ,5).— Descriptions of quack grass and the 

 Canada thistle are followed by directions for their eradication. 



HORTICULTURE. 



Report of the horticulturist, C. F. Kinman (Porto Rico Sta. Rpt. 1909, pp. 

 19-23). — A brief report on the condition of temperate climate and citrus fruits, 

 mangoes, avocados, and miscellaneous plants being tested, together with an 

 outline of the experiments being conducted with fruits and vegetables. 



The apple, pear, plum, and persimmon trees in the station orchards have made 

 some growth, but it is unlikely that these fruits will prove of commercial value 

 in that locality. 



Trees of improved varieties of mangoes set out in 1904 are now from IS to 20 

 ft. tall and it appears that they will prove to be an excellent windbreak for the 

 citrus orchards. The avocados in the station orchard grow fairly well until 



