Forage in vitro rumen undigested NDF, corrected for ash (uNDFom, % of DM) is thought related to gut fill and dry matter intake in dairy cattle, however assay repeatability between laboratories, is unknown for long-term digestion periods. The objective here was to evaluate uNDFom repeatability for extended time points between 2 forage analysis laboratories. European grass silages were selected based upon NIR predictions to range in aNDF and uNDFom. Samples (n = 27) were dried, ground (1mm), mixed and split for analysis at both Rock River Laboratory (RRL) and Cornell University (CU) forage analysis laboratory for in vitro rumen uNDFom at 30, 120 and 240 h. Cornell University utilized the Raffrenato et al. (2018) technique and RRL used a modification of this technique, including Goeser and Combs (2009) rumen fluid preparation, re-inoculation for 240h time period (at 120h), and using an Ahlstrom 141 glass fiber filter (3.1-µm pore size) to filter samples post digestion and refluxing. Samples were digested in triplicate, within in vitro run, at each time point, and in 2 runs at CU and 1 run at RRL. The single run was chosen at RRL to mirror commercial feed analysis procedures. Following sample analyses, uNDFom was related back to laboratory, time point, and the interaction using the Fit Model function in JMP v14.0. Laboratory and time point were treated as fixed effects. The interaction was not significant and was removed. Following linear modeling, uNDFom was related to time (P < 0.0001) but not laboratory (P > 0.4). Model adjusted R-square was 0.23 and RMSE was 6.9. The least squares means (lsmean) for European grass silage uNDFom at each digestion time point (% of DM) were 18.9, 11.0, and 10.3 for 30, 120 and 240h, respectively, across both labs. The lsmean uNDFom for each laboratory were 13.8 and 13.0 for CU and RRL, respectively. Based on these observations for selected European grass silages, uNDF measures appear repeatable between 2 different laboratories. However further comparisons, with alternative forage and feed types, are warranted to better understand repeatability for long-term incubated samples.