May 1, 1936 Spores of BocUlus Larvae in Honey 703 



larvae in 1 liter of sugar sirup, or less than 50,000 spores per cubic 

 centimeter. Of 11 colonies fed 50,000 spores per cubic centimeter, 2 

 developed disease and 9 remained healthy; of 6 colonies fed 75,000 

 per cubic centimeter, 3 developed positive disease and I probable 

 disease, and 2 remained healthy; of 6 colonies fed 100,000 per cubic 

 centimeter, 2 were positive, 1 probable, and 3 remained healthy; of 

 4 colonies fed 200,000 spores per cubic centimeter, 3 were positive 

 and 1 probable. Thus it was assumed that 50,000 spores per cubic 

 centimeter of sirup could be . considered the critical number or 

 minimum infectious dose of spores that will produce disease, when 

 1 liter is used as the unit volume to be fed. 



Since the foregoing analysis of the data indicates, by the method of 

 estimating used, that the actual mean number of spores per field falls 

 well withm the limits of permissible error for the estimated actual 

 means (±3 times the standard error of estimate), the most probable 

 value for such a mean for use in determining the number of spores per 

 cubic centimeter of an unknown sample is the actual mean number of 

 spores per field determined by counting 30 fields each from stained 

 smears from two centrifuged sediments of this^ample. If the formula 

 X=1.1228F+0.2034 is used to estimate X, the most probable 

 theoretical number of spores that should have been recovered, when 

 Y represents the actual mean number of spores per field, and if this 

 value is then multiplied by 32,884, the most probable number of 

 spores per cubic centimeter in the unknown sample can be calculated. 

 Applying the limits of error for X,±3 times the standard error of esti- 

 mate, or ±0.4374 spore, and carrying it through into the second for- 

 mula will give the possible range in which the number of spores per 

 cubic centimeter might fall within the precision of the method. 



Further work is in progress to determine whether the same accuracy 

 will be obtained by counting a smaller number of fields to obtain the 

 mean number of spores per field from a larger number of smears from 

 sediments. 



Since in the experimental work the samples of known spore content 

 contained approximately round numbers of spores — multiples of 

 50,000 — ^it probably would be sufficiently accurate to designate the 

 number of spores as the nearest multiple of 50,000 to the actual figures 

 derived from the formulas. When using the limits of error 0± 0.4374 

 spore per field, for the estimated mean number of spores per field, it 

 wiU be found that for numbers below 100,000 there will be some 

 overlapping between 10,000-spore increments, and the value will 

 have to be expressed approximately (for example, the honey contains 

 between 40,000 and 60,000 spores per cubic centimeter); neverthless 

 the honey can still be designated either as dangerous or as not 

 dangerous. 



SUMMARY 



Previous work on the qualitative demonstration of the presence or 

 absence of spores of Bacillus larvae in honey that has been in contact 

 with American foulbrood has been followed by the development of a 

 quantitative method for determining the approximate number of 

 spores per cubic centimeter in such honey. The method is represented 

 by the formula 



Number of spores per cubic centimeter= ^^ 



