| Creator |
|
| Language |
|
| Publisher |
|
|
|
|
|
|
|
| Date |
|
| Source Title |
|
| Vol |
|
| First Page |
|
| Last Page |
|
| Publication Type |
|
| Access Rights |
|
| Crossref DOI |
|
| Related DOI |
|
| Related URI |
|
| Relation |
|
| Abstract |
The effect of using eleven subsets of characters on the clustering method was tested based on 85 taxa or OTUs of the Japanese Andrenid bees (Hymenoptera, Andrenidae). The eleven character subsets, whi...ch consisted of total (Nos.=130), randomly selected (100, 70, and 40), head-thoracic (40 and 58), total structural (40 and 82), PCA produced (21- weighted and 21-unweighted) and key characters (21), were analyzed by cluster analysis based on distance coefficients. The results were evaluated and discussed at the subgeneric level of the genus Andrena. The two phenograms derived from the 100 and 130 character subsets were similar to each other. The two PCA produced subsets not necessarily resulted in enough groupings, The 21 key character subset produced different clusters far from that derived from the total characters.show more
|
Export Mendeley
29