In a recent article, Tang et al. [1] proposed a general model explaining the switching and negative differential resistance (NDR) effect observed in a multitude of different molecular and organic thin-film device configurations. Tang et al. argued that switching and NDR are general phenomena originating from nucleation of evaporated-metal nanoislands formed in crevices created in the vicinity of dust particles and/or other impurities on the bottom electrode. If these nanoislands form in a special four-sphere configuration, essentially a 2D ring structure, anomalous current-voltage characteristics are predicted. [2] In order to obtain the correct order of magnitude for the experimentally measured current density, Tang et al. argued that there have to be contributions from 10 4 independent four-ring island arrays arising at the 10 4 dust and/or impurity sites having crevices. We do acknowledge the possibility of metal-island formation in crevices around possible dust particles. However, we do not agree with this being the general solution to switching and NDR observed in many types of materials. We have performed experiments where we have observed that, upon careful evaporation of aluminum with standardized parameters, such as the evaporation rate, the distance of the mask from the electrode, and the pressure during evaporation, the effect of pinholes can be minimized and a polystyrene (PS) film between two aluminum electrodes can act as an insulator. By keeping all the evaporation parameters fixed, we have systematically introduced a small fraction of fullerenes (C60) in the PS matrix (the details of the experimental procedure have been described previously [3] ) and have obtained completely different electrical behaviors. The films made from 1 wt % of C60 in PS were as insulating as the PS-only device (Fig. 1). But, when we made a device from a solution with 5 wt % of C60 in PS, we found switching and NDR in an all-organic active layer, as clearly seen in Figure 1. However, the switching behavior has only a 40 % reproducibility between different devices and is not an additive effect, as it should be if the suggested model were correct.[1] We either observe or do not observe the effect in different cells on the same substrate. We have measured a large set of substrates and, in general, we found that four out of ten of our 5 wt % devices exhibit switching, as has also been reported for the intermediatemetal-based devices.[4]Upon increasing the amount of C60 in the devices further, the switching disappears and only a systematic NDR is observed. The reproducibility of the NDR is 100 % for all the measured devices. We have also observed switching and NDR in devices made without an evaporated top electrode, where we have used solution-cast poly(3,4-ethylenedioxythiophene):poly(styrene sulfonate) (PEDOT:PSS) as the top electrode. This further justifies the fact that the observed switching is not only due to the metal inclusion from evaporation.We have tried to find possible proof for four-ring nanoisland arrays in our C...