Enantiomerically pure allylboronic esters 1 + 2 with a stereogenic center a to the boron moiety can be obtained by a sigmatropic rearrangement of boron containing allyl alcohols. Allyl additions with the new reagents are highly selective, which was shown via the direct measurement of the diastereoisomeric ratio of the intermediates 5 + 6 by characteristic NMR chemical shifts. The observations are not limited to ester containing reagents, but holds also true for hydrocarbon side-chains (e.g. in 11 + 12) that were readily obtained by reducing the ester.One of the key reactions in organic synthesis is the allyl addition, especially using allylboronic esters; regularly homoallylic alcohols are conveniently formed in high yield and enantiomeric excess. 1-6 Reagents having a stereogenic center in the position a to the boronic ester are less often used since they are more difficult to prepare in enantiomerically pure form. 7-14 We have recently demonstrated that highly stable reagents of the general type 1 or 2 are readily available via a Johnson rearrangement of the corresponding boron-substituted allyl alcohols. [15][16][17] The derivatives are easy to handle and store, and add highly selectively to a number of aldehydes giving either homoallylic alcohol 3 or 4 with the enantiomeric excess ranging from 92 to >99% (Scheme 1). The formation of the Z-double bond and the configuration of the newly formed stereogenic centers were unambiguously proven by means of chemical correlation. The results could also be rationalized by a transition state as shown in Scheme 1 -the substituent in position a to boron is preferentially axial -which is in full agreement with a previous report by Hofmann and Weidmann. 7 A drawback of the procedure was the fact that the enantiomeric excess of homoallylic alcohols was regularly determined -as it is common practice -by forming diastereoisomeric Mosher ester 18,19 (when direct methods fail) and thus only indirectly establishing the stereochemical outcome of the transformation. Obviously, there are two problems associated with the approach: A diastereomeric discrimination of the ester formation must be ruled out and, more importantly, the hydrolysis of the intermediate boric esters (in our case 5 and 6) must occur with similar rates. In many cases this might not be a problem; however, occasionally we observed a rate difference that would lead to incorrect results, even when direct methods to determine the enantiomeric excess were used. An obvious solution to the problem would be the utilization of the formed diastereoisomeric intermediates 5 and 6 that should show distinct differences in their NMR spectra. Hence we decided to start a NMR investigation of the reaction before work-up and especially chromatographic separation.First, we investigated the most simple derivatives 5a and 6a ( Figure 1) and found that almost all signals in the proton NMR (500 MHz) show distinct differences in the Scheme 1 Allyl additions of 1 or 2.