Creating document-feature matrix from already stemmed token fails

[AdroMine]

Using quanteda 0.9.5.8 with both R 3.2.3 on Windows 7 x64. If I try to create a dfm from already stemmed tokens (stemmed using wordstem() ), I get an error. Using stem within the dfm command does work however.

Reproducible Code -->

corp <- corpus(inaugTexts)
tokens <- tokenize(corp)
stemTokens <- wordstem(tokens)
dfm1 <- dfm(stemTokens)

Creating a dfm from a tokenizedTexts object ...
   ... indexing documents: 57 documents
   ... indexing features: 6,257 feature types
Error in (function (cl, name, valueClass)  : 
  assignment of an object of class “NULL” is not valid for @‘ngrams’ in an object of class “dfmSparse”; is(value, "integer") is not TRUE