ON THE MERITS OF IRON AND WOOD f OR ROOFS 81 



cal experience that water will not circulate beyond the point 

 where there is an accumulation of air, and the more powerful 

 the attempts made to remove the obstruction by increasing the 

 strength of the fire, &c, the more likely is the apparatus to work 

 improperly, and to cause an overflow of water in the supply cis- 

 tern. Therefore, as soon as it is discovered by the gardener or 

 person having the management of the fire, that the water does 

 not circidate regularly, he should trace the water by its warmth 

 along the pipe to the place where he finds the metal cold ; and 

 then in the next bind, or angle, should he not find an air tap, I 

 should recommend him to procure a blacksmiths drill and to have 

 a hole made in the pipe, when he will find the air to pass off 

 rapidly and the water to follow instantly. Then, should he not 

 be prepared with an air pipe, a small wooden plug would suffice 

 until an opportunity offered to fix one properly ; as in all proba- 

 bility it might be many months and perhaps years: if the cistern 

 be carefully and continually attended to, before such an accident 

 coidd accur again at that particular joint or bend, I woidd, how- 

 ever, strongly recommend that in every apparatus ample pro- 

 vision should be made for the escape of the air, at every bend 

 where it is likely to collect or lodge, for I have witnessed during 

 the time I had the management of six systems of hot water, that 

 from some unknown cause, an apparatus which had worked 

 properly for one or two years, would suddenly get out of order, 

 when, on tracing the pipe as before described, as far as I found 

 it warm, I have then, on drilling a hole at the first turn or bend 

 where the pipe began to feel cold, found an accumulation of hy- 

 drogen carbonic acid gas, the heaviest of all the gasses lodged 

 in the angle, and as soon as this was allowed to escape, the appa- 

 ratus worked as regularly as usual. To remove this difficulty 

 which to persons unacquainted with the cause of the obstruction, 

 would appear formidable, nay almost insurmountable, not more 

 than ten minutes space was sacrificed ; and on interrogating the 

 man who had the management, as to whether he had allowed the 

 cistern to fall below the proper level, I discovered that the de- 

 rangement had been caused through his negligence and inatten- 

 tion, in having suffered the water to sink below the level of the 

 top pipes, which of course left a vacuum for tins foul air. I 

 would therefore advise all persons, when not using the apparatus, 

 < ither to draw the whole of the water off, or to keep the cistern 

 11 in use ; this precaution will prevent the air from 



