Mann has adopted precisely the opposite course. He has, first of all, dealt with the total column, then with the Church, and he has lastly subtracted the one set of results from the other. The consequence is he has assigned no more than 274 of the dateless meeting-houses to the period before 1801. The total number he has distributed amongst the first three periods is only 737, whereas he has divided no fewer than 1,691 amongst the last three. It need scarcely be said that all the probabilities would be all in favour of reversing the process. At the outset, therefore, Mr. Mann’s estimate comes before us under circumstances of extreme suspicion; but, granting, for the sake of argument, that his distribution of the existing meeting-houses were correct, it must be obvious that any inference from dates would be preposterous unless we could be certain that there were no buildings in existence at the earlier periods, other than those included in the table. It has been seen that Mr. Mann has not overlooked this circumstance. He admits that the small number assigned to 1801 “seems to prove that many dissenters’ buildings existing in former years have since become disused or have been replaced by others;” but no one would suspect from this statement the vast number of these disused buildings. Take, for example, the case of Nottingham. From Mr. Wylie’s local history it would appear that of the 29 meeting-houses returned to the Census Office, only six dated back to the commencement of the present century. In other words, dissent in Nottingham, on Mr. Mann’s hypothesis, all but quintupled itself during the 50 years. In point of fact, however, there were, not six, but thirteen or fourteen, dissenting congregations in 1801, and probably several more whose “memorial has perished with them.” The absurdity of the Census estimate may be still further illustrated by a reference once more to Tables 6 and 14. Those tables are to Mr. Mann’s calculation not very different from the proof of an addition sum. If his estimate were right they would agree with Tables 5 and 13; but instead of doing so, they lead to the following astounding results:—In 1851, there were in the Meeting Houses. Sittings. Average Sittings. Large town 6,129 2,131,515 347 each. districts Residue of country 14,261 2,763,133 193 „ This is, of course, quite correct. But now see what the tables say of 1801— Meeting Houses. Sittings. Average Sittings. Large town 1,337 258,220 193 each. districts Residue of country 2,634 781,218 330 „