142 LETTER-FILES OF S. W. JOHNSON 



act themselves, might find a mercenary motive in his sustain- 

 ing a school at his own expense for 200 poor children in Lon- 

 don, or for his oiJering the cottagers of this village 600£ to 

 improve their cottages with so soon as they raised 1/2 that 

 sum, or for his building them a library to try and elevate the 

 condition of the laboring men. . . . 



They have supplied me with about $500. worth of apparatus, 

 and we have been doing up the subject on a scale unprece- 

 dented. I have 12 glass shades 3 1/2 ft. high and 9 1/2 inches 

 in diameter, and under them, entirely isolated from the air, 

 we are growing plants and forcing NHg-free air into them 

 daily. Day before yesterday we had a visit from Prof. 

 Graham, Noad (author of analysis), Bolard (French) and 

 5 or 6 others. Our results indicate a confirmation of Bous- 

 singault. The evidence accumulates. ... I asked Bolard how 

 he accounted for Ville's plants growing as they did. He 

 answered "II a ajoute sans doute." He said nobody in Paris 

 trusted them. Indeed I have not yet decided whether to treat 

 Yille as though he didn't exist (silent contempt) or to expose 

 him. Boussingault said in a letter the other day to Dr. Gilbert 

 "from my own experiments I am fully persuaded that plants 

 don't assimilate N, yet," continued he, "so great is my confi- 

 dence in the Rothamsted experiments that should you get 

 an appreciable increase of N, I shall modify my views." . . . 

 Yours most truly, E. Pugh. 



In the fall of 1859, urged by a patriotic sense of duty, 

 Evan Pugh accepted the presidency of the Pennsyl- 

 vania State Agricultural College and returned to 

 America, where he employed his talents and his 

 knowledge in establishing his college on a broad and 

 enduring basis ; with such success that at the time of 

 his early death in 1864, he was recognized as one of 

 the most able of the men then engaged in the advocacy 

 of scientific agricultural education in this country. 



