The Sekonic 308 has a 40 degrees angle when measuring reflected light. That's the same coverage of a 50mm lens but the Nikon is center weighted giving more weight to the center thus their readings are a bit higher. The FE#2 is different from #1 about 1/2 stop which is normal. When measuring incident light the Sekonic read 2 stop lower because the wall is about 80% reflectance instead of 18%. Pointing the reflected light meter toward lens or pointing the dome toward the subject when in incident mode don't make good reading. So I have to concluded that there was something wrong with the 35mm lens. Having wider angle should result in somewhat lower reading but not as much as the OP experienced.