Introduction

Light is a key environmental factor that drives many aspects of plant growth and development1. Phototropism, the reorientation of growth towards light, is one of the most important adaptive processes2. Many results have been acquired for phototropism in a variety of aspects since Charles Darwin3 published “The Power of Movement in Plants” e.g.1,2,4,5,6,7,8,9,10,11. Most growth ring studies focused on the dendrochronology as well as its utility for palaeoecologic and palaeoclimatic investigations in deep time e.g.12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33. However, little attention has been paid to the trunk phototropism represented by asymmetric growth of tree rings in response to a directed light source. In recent years, increased data have been accumulated for such eccentricity phototropism in both living and fossil tree ring observations, especially in some well-preserved individual fossil trunks34,35. However, such evidence is still very limited and poorly known in fossil record and the relationship between ring phototropism and rotation of block is undetermined. Here we report our recent systematic field surveys and investigation results on both living trees and in situ fossil wood from the North China Block, including 253 living trees from Bei**g and Jilin Provinces, and 7 fossil in situ trunks from the Jurassic Tiaojishan and Tuchengzi formations in Liaoning and Bei**g regions (Supplementary Information S1).

As eccentricity refers to the directional measurement of the longest distance from the pith to the outermost growth ring of an in situ tree trunk (southwest direction, when a tree with two largest growing directions), we thus use tree ring eccentricity as a proxy to determine the general block rotation. In addition, our palaeomagnatic data from the Tiaojishan Formation in Beipiao of Liaoning Province also provide support for the fossil data interpretation, indicating that the North China Block had rotated clockwise since the Middle to Late Jurassic transition (Supplementary Information S1).

Results

Phototropism in extant tree trunks

The shape, dominant eccentricities and other directions from the pith to the largest growing part (when a tree trunk with two largest growing parts) of living tree trunks were observed within the latitude ranges from 39°59.6′N to 43°15′N in northern China region. We measured 35 trunks at the Wofusi locality in ** of living wood

All the wood trunks in **angshan of Bei**g, Huadian of Jilin Provinces were perpendicular to the horizon. Ruler was used to measure the outermost radius. Compass was applied to measure the eccentricity, e.g. the direction (southwest direction, when a tree has two largest growing parts) from the pith to the largest growing part. GPS was used to locate the precise positions. In order to exclude influencing factors other than light stimuli, tree trunks found in flat ground without any shadowing obstacles were selected as the ideal data. If the data were influenced by other factors, it was crucial to eliminate the interference factors in field study. In the common sense, the southwest direction is probably the phototropism direction; other directions are formed by the lateral factors.

We mapped the numbers of living tree data points (253 data) collected in each sector into a radar figure, in which the 360-degree circle was divided evenly into 60 sectors, each sector represents 6 degrees, and the data is arranged by angle in ascending order and sorted into the corresponding sectors.

Field collection of in situ petrified wood

In situ preserved petrified wood in two sites were discovered during the fieldwork. Two sites are located in Beipiao of Liaoning for the Tiaojishan Formation and in Bei**g for the Tuchengzi Formation. In the Tiaojishan Formation, three stumps were found for measuring the eccentricity. In the Tuchengzi Formation, four stumps were chosen to indicate the phototropism. All the petrified wood trunks are vertical to the bedding plane to ensure in situ status. Ruler, compass and GPS units were used when investigating fossil wood in the field. We measured the largest off-centre data of fossil wood, which is from the pith to the largest outermost of the trunk. In the field work, some fossil trunks have two largest growing directions, i.e. No. TJS 1 (233° and 305°) and No TCZ 3 (236° and 44°). The southwest directions (233° and 236°) are thus the direction of eccentricity, while the other directions (305° and 44°) may due to other influencing factors, i.e. shaded with other trees. For the in situ fossil trunks preserved in the open and flat environment and far away from the water sources39,40,41, the influencing factor, i.e. gravity and near water sources are excluded.

During the field survey, we emphasized that both living and fossil trees may have a single, non-hollow trunk, and exhibit annual growth cycles. Structural changes caused by disease or damage by insects were not observed. Because the difference of tree species types does not affect heavily on the phototropism of living wood50,51, therefore we followed this principle and did not distinguish their taxonomic position. In our study, all the fossil wood trunks are represented by gymnosperms. Living trees in **angshan of Bei**g are angiosperms, whereas trees in Huadian of Jilin are 80% gymnosperms (conifers), and 20% are angiosperms.

All the growth rings in both living and fossil trees are larger than 30 circles, which means the trees are all older than 30 years. The average height of in situ preserved trees in the Tiaojishan Formation is about 25 m52.

Palaeomagnetic data

We collected ten sites (approximately 100 samples) for palaeomagnetic analysis from the Tiaojishan Formation in Beipiao Basin, very close to the in situ preserved petrified wood samples. The occurrence of the stratum was measured on the intercalated sandstone layer in volcanic tuff (Supplementary Information S3). The strike and dip of the stratum is 42° and 14°, respectively. In total, material from ten sites were sampled using a gasoline-powered drill, and approximately ten oriented samples were collected from each site.

The samples were cut into cylinders 2.2 cm long for subsequent palaeomagnetic analysis. All samples underwent stepwise thermal demagnetization up to 680 °C that was performed with an ASC TD-48 thermal demagnetizer with an internal residual field of <10 nT. The demagnetization temperature intervals were generally large (40–50 °C) in the low-temperature part and smaller (20–30 °C) at higher temperatures. Remnant magnetizations were measured using a 2G-755R cryogenic magnetometer and a JR-6 spinner magnetometer. All measurements were carried out in a shielded room with residual fields of <300 nT at the Key Laboratory of Palaeomagnetism and Tectonic Reconstruction of the Ministry of Land and Resources, Institute of Geomechanics, Chinese Academy of Geological Science in Bei**g. Magnetization directions were determined by principal component analysis53 or remagnetisation circle analysis54. The average palaeomagnetic direction was counted with Fisher statistics55 or the mixed mean of the unit vectors and remagnetisation great circles56. The computer program Kirsch developed by Enkin57 and PaleoMac developed by Cogné58 were used to analyse the palaeomagnetic data.