BUNSEN MEMORIAL LECTURE. 029 



chief aim of the research, the preparation of pure rhodium and its 

 compounds. 



In the course of these experiments Runsen met with a sinirular and 

 unexplained accident, which fortunately had no serious c-onsequouces. 

 With reference to this he writes to me as follows: 



"It is still difficult for me to write, as my hands are not f|uite healed, 

 but I can not longer delay replying to your sympathetic letter, as 1 

 fear you may be uneasy about me. The cause' of the exjjlosicjn is to 

 me still quite inexplicable. I had prepared about a pound of the mixed 

 metals rhodium and iridium by zinc reduction, and had diicd the 

 powder at 100° in a water bath, when, on liohtly touciiinn- the tiiielv 

 divided metal, which was not quite cold, with my tinjicr. the wiiolc 

 suddenly exploded with the eneroy of rammed-in ounpowder. This 

 is all the more puzzling-, as I have <)ft(Mi rubbed a powder of the same 

 metals violentl}' in a mortar in similar ciuantities without any explosion 

 occurring-. I have also heated similar preparations to a redness in \acuo 

 without any gas, and certainly without a trace of hydrogen, hcing 

 evolved. My left hand, with the first finger of which I touched the 

 mass, saved m}^ eyes, as my face and eyes were only superficially burned 

 by the flames which penetrated through my fingers. My eyes are, 

 with the exception of singed eyebrows and eyelaslies. unhurt, and so 

 the explosion will luckily leave behind no serious traces." 



In the preceding communication on the platinum metals Bun-en 

 first describes the well-known filter pump which now bears his name. 

 But in a later publication (Annalen, 1868 (148), 269) he gives further 

 particulars of its construction and use. These are so well known 

 that it is only necessary to say that it is, in fact, a Spi-engel pump 

 in which a column of 28 inches of mercury is replaced by one of 

 32 feet of water. In this way a flow of water down a pipe of the 

 above length produces a vacuum perfect up to the limit of tensicm of 

 the aqueous vapor, and under the diminished pressures thus l)r()ught 

 about all the processes of filtration and of the washing of precipitates 

 can be carried out with nuich greater rapidity and perfection than is 

 the case when working under the ordinary atmospheric pressure'. I lere, 

 as in all his published work, Bunsen is precise and exact. To show 

 the timesaving value of the process he precipitates two equal vohnnes 

 of chromium sesquichloride solution of known strength by ammonia: 

 the one portion he treats in the ordinary way, the otliei- by the tilter- 

 pump method, whereby he demonstrates that, treated by the latter 

 process, the precipitate is completely washcMl in one-thirteenth part ..f 

 the time needed bv the old plan, while only one-fiftieth of the volume 

 of wash water is required. Such filter pumps, furnished with mer.ury 

 pressure gauges, are now found in every well-fitted laboratory. 



A somewhat simpler form of filter pump, first des.-ribed by Tiecard 

 (Zeit. anal. Chem., 1865 (4), 45), is, however, now als,» very gen.'rally 

 emploved. This consists of a short glass tube attach.-d to the water 

 tap with an inner jet for the water and an outer air tube, the rapid 



