"Once the fruit had matured on the plants in
the field, a single harvest of 25 random
mature fruits from at least 10 plants in each
replication was bulked. After harvest, the
sample was dried and ground. The extraction
of the capsaicinoids and the estimation of
capsaicinoid amounts followed the highperformance
liquid chromatography (HPLC)
procedures for the short run method as
described by Collins et al. (1995). The HPLC
data were converted from parts per million to
SHU by multiplying the parts per million
by 16. Samples of the ground fruits were
also sent to two commercial laboratories,
Southwest Bio-Laboratories (Las Cruces,
N.M.) and Ag-Biotech (Gilroy, Calif.), for
heat level analysis to validate our results."
"The growing season in 2005 was favorable
for the production of fruits on all three
chile pepper cultivars. The environment is
known to affect the heat level of chile pepper
cultivars (Harvell and Bosland, 1997).
Having a replicated field trial with standard
control cultivars allows for a better comparison
of heat levels among cultivars. The
HPLC analysis revealed that orange habanero
had a mean heat level of 357,729 SHUs,
which is in the range normally seen for this
cultivar in Las Cruces, N.M. (Table 2). The
results of the analysis for ‘Bhut Jolokia’
indicated that it possessed an extremely high
heat level, 1,001,304 SHUs, whereas ‘Red
Savina’ recorded a heat level of 248,556
SHUs. Independent tests confirmed this high
level of heat for ‘Bhut Jolokia’ with 927,199
SHUs and 879,953 SHUs from Southwest
Bio-Laboratories and Ag-Biotech, respectively."
They used the maximum test answer but not the maximum result, but you will note that they used THEIR OWN result. This is more important than the fact that they used the HIGHEST result since they can do multiple tests. I believe it was in fact multiple samples that were tested to get that result. Each sample used 25 pods randomly selected. The paper was a bit unclear as to whether they had more than one sample tested at the CPI lab (it seems like they did, they used the word mean for their samples), but they only sent one sample to each of the 3rd party labs. At the very least it was a mean result of 25 pods from various plants, so yes, it is a mean. I believe the lab at the CPI used multiple samples and had the results which they compared to other fruits so they used their own source as a more reliable estimate. This is why I like the CPI testing, they don't use one result that got spit back from a single sample and a single lab. Of course then you have to answer what the 'max' 'min' and 'mean' values mean here, whether they are results from multiple labs, and how many pods went into each of these samples. If you can find more specific information I'd like to see it. Thanks guys.
Edited by mrz1988, 07 February 2012 - 01:26 PM.